categories
string | doi
string | id
string | year
float64 | venue
string | link
string | updated
string | published
string | title
string | abstract
string | authors
sequence |
|---|---|---|---|---|---|---|---|---|---|---|
null | null | 2406.09199 | null | null | http://arxiv.org/pdf/2406.09199v1 | 2024-06-13T14:56:52Z | 2024-06-13T14:56:52Z | Precise analysis of ridge interpolators under heavy correlations -- a
Random Duality Theory view | We consider fully row/column-correlated linear regression models and study several classical estimators (including minimum norm interpolators (GLS), ordinary least squares (LS), and ridge regressors). We show that emph{Random Duality Theory} (RDT) can be utilized to obtain precise closed form characterizations of all estimators related optimizing quantities of interest, including the emph{prediction risk} (testing or generalization error). On a qualitative level out results recover the risk's well known non-monotonic (so-called double-descent) behavior as the number of features/sample size ratio increases. On a quantitative level, our closed form results show how the risk explicitly depends on all key model parameters, including the problem dimensions and covariance matrices. Moreover, a special case of our results, obtained when intra-sample (or time-series) correlations are not present, precisely match the corresponding ones obtained via spectral methods in [6,16,17,24]. | [
"['Mihailo Stojnic']"
] |
null | null | 2406.09206 | null | null | http://arxiv.org/pdf/2406.09206v1 | 2024-06-13T15:06:11Z | 2024-06-13T15:06:11Z | Self-Training for Sample-Efficient Active Learning for Text
Classification with Pre-Trained Language Models | Active learning is an iterative labeling process that is used to obtain a small labeled subset, despite the absence of labeled data, thereby enabling to train a model for supervised tasks such as text classification. While active learning has made considerable progress in recent years due to improvements provided by pre-trained language models, there is untapped potential in the often neglected unlabeled portion of the data, although it is available in considerably larger quantities than the usually small set of labeled data. Here we investigate how self-training, a semi-supervised approach where a model is used to obtain pseudo-labels from the unlabeled data, can be used to improve the efficiency of active learning for text classification. Starting with an extensive reproduction of four previous self-training approaches, some of which are evaluated for the first time in the context of active learning or natural language processing, we devise HAST, a new and effective self-training strategy, which is evaluated on four text classification benchmarks, on which it outperforms the reproduced self-training approaches and reaches classification results comparable to previous experiments for three out of four datasets, using only 25% of the data. | [
"['Christopher Schröder' 'Gerhard Heyer']"
] |
null | null | 2406.09207 | null | null | http://arxiv.org/pdf/2406.09207v1 | 2024-06-13T15:08:44Z | 2024-06-13T15:08:44Z | Investigating potential causes of Sepsis with Bayesian network structure
learning | Sepsis is a life-threatening and serious global health issue. This study combines knowledge with available hospital data to investigate the potential causes of Sepsis that can be affected by policy decisions. We investigate the underlying causal structure of this problem by combining clinical expertise with score-based, constraint-based, and hybrid structure learning algorithms. A novel approach to model averaging and knowledge-based constraints was implemented to arrive at a consensus structure for causal inference. The structure learning process highlighted the importance of exploring data-driven approaches alongside clinical expertise. This includes discovering unexpected, although reasonable, relationships from a clinical perspective. Hypothetical interventions on Chronic Obstructive Pulmonary Disease, Alcohol dependence, and Diabetes suggest that the presence of any of these risk factors in patients increases the likelihood of Sepsis. This finding, alongside measuring the effect of these risk factors on Sepsis, has potential policy implications. Recognising the importance of prediction in improving Sepsis related health outcomes, the model built is also assessed in its ability to predict Sepsis. The predictions generated by the consensus model were assessed for their accuracy, sensitivity, and specificity. These three indicators all had results around 70%, and the AUC was 80%, which means the causal structure of the model is reasonably accurate given that the models were trained on data available for commissioning purposes only. | [
"['Bruno Petrungaro' 'Neville K. Kitson' 'Anthony C. Constantinou']"
] |
null | null | 2406.09241 | null | null | http://arxiv.org/pdf/2406.09241v1 | 2024-06-13T15:44:23Z | 2024-06-13T15:44:23Z | What is the long-run distribution of stochastic gradient descent? A
large deviations analysis | In this paper, we examine the long-run distribution of stochastic gradient descent (SGD) in general, non-convex problems. Specifically, we seek to understand which regions of the problem's state space are more likely to be visited by SGD, and by how much. Using an approach based on the theory of large deviations and randomly perturbed dynamical systems, we show that the long-run distribution of SGD resembles the Boltzmann-Gibbs distribution of equilibrium thermodynamics with temperature equal to the method's step-size and energy levels determined by the problem's objective and the statistics of the noise. In particular, we show that, in the long run, (a) the problem's critical region is visited exponentially more often than any non-critical region; (b) the iterates of SGD are exponentially concentrated around the problem's minimum energy state (which does not always coincide with the global minimum of the objective); (c) all other connected components of critical points are visited with frequency that is exponentially proportional to their energy level; and, finally (d) any component of local maximizers or saddle points is "dominated" by a component of local minimizers which is visited exponentially more often. | [
"['Waïss Azizian' 'Franck Iutzeler' 'Jérôme Malick'\n 'Panayotis Mertikopoulos']"
] |
null | null | 2406.09246 | null | null | http://arxiv.org/pdf/2406.09246v1 | 2024-06-13T15:46:55Z | 2024-06-13T15:46:55Z | OpenVLA: An Open-Source Vision-Language-Action Model | Large policies pretrained on a combination of Internet-scale vision-language data and diverse robot demonstrations have the potential to change how we teach robots new skills: rather than training new behaviors from scratch, we can fine-tune such vision-language-action (VLA) models to obtain robust, generalizable policies for visuomotor control. Yet, widespread adoption of VLAs for robotics has been challenging as 1) existing VLAs are largely closed and inaccessible to the public, and 2) prior work fails to explore methods for efficiently fine-tuning VLAs for new tasks, a key component for adoption. Addressing these challenges, we introduce OpenVLA, a 7B-parameter open-source VLA trained on a diverse collection of 970k real-world robot demonstrations. OpenVLA builds on a Llama 2 language model combined with a visual encoder that fuses pretrained features from DINOv2 and SigLIP. As a product of the added data diversity and new model components, OpenVLA demonstrates strong results for generalist manipulation, outperforming closed models such as RT-2-X (55B) by 16.5% in absolute task success rate across 29 tasks and multiple robot embodiments, with 7x fewer parameters. We further show that we can effectively fine-tune OpenVLA for new settings, with especially strong generalization results in multi-task environments involving multiple objects and strong language grounding abilities, and outperform expressive from-scratch imitation learning methods such as Diffusion Policy by 20.4%. We also explore compute efficiency; as a separate contribution, we show that OpenVLA can be fine-tuned on consumer GPUs via modern low-rank adaptation methods and served efficiently via quantization without a hit to downstream success rate. Finally, we release model checkpoints, fine-tuning notebooks, and our PyTorch codebase with built-in support for training VLAs at scale on Open X-Embodiment datasets. | [
"['Moo Jin Kim' 'Karl Pertsch' 'Siddharth Karamcheti' 'Ted Xiao'\n 'Ashwin Balakrishna' 'Suraj Nair' 'Rafael Rafailov' 'Ethan Foster'\n 'Grace Lam' 'Pannag Sanketi' 'Quan Vuong' 'Thomas Kollar'\n 'Benjamin Burchfiel' 'Russ Tedrake' 'Dorsa Sadigh' 'Sergey Levine'\n 'Percy Liang' 'Chelsea Finn']"
] |
null | null | 2406.09250 | null | null | http://arxiv.org/pdf/2406.09250v1 | 2024-06-13T15:55:04Z | 2024-06-13T15:55:04Z | MirrorCheck: Efficient Adversarial Defense for Vision-Language Models | Vision-Language Models (VLMs) are becoming increasingly vulnerable to adversarial attacks as various novel attack strategies are being proposed against these models. While existing defenses excel in unimodal contexts, they currently fall short in safeguarding VLMs against adversarial threats. To mitigate this vulnerability, we propose a novel, yet elegantly simple approach for detecting adversarial samples in VLMs. Our method leverages Text-to-Image (T2I) models to generate images based on captions produced by target VLMs. Subsequently, we calculate the similarities of the embeddings of both input and generated images in the feature space to identify adversarial samples. Empirical evaluations conducted on different datasets validate the efficacy of our approach, outperforming baseline methods adapted from image classification domains. Furthermore, we extend our methodology to classification tasks, showcasing its adaptability and model-agnostic nature. Theoretical analyses and empirical findings also show the resilience of our approach against adaptive attacks, positioning it as an excellent defense mechanism for real-world deployment against adversarial threats. | [
"['Samar Fares' 'Klea Ziu' 'Toluwani Aremu' 'Nikita Durasov' 'Martin Takáč'\n 'Pascal Fua' 'Karthik Nandakumar' 'Ivan Laptev']"
] |
null | null | 2406.09253 | null | null | http://arxiv.org/pdf/2406.09253v1 | 2024-06-13T15:56:55Z | 2024-06-13T15:56:55Z | Deep Sketched Output Kernel Regression for Structured Prediction | By leveraging the kernel trick in the output space, kernel-induced losses provide a principled way to define structured output prediction tasks for a wide variety of output modalities. In particular, they have been successfully used in the context of surrogate non-parametric regression, where the kernel trick is typically exploited in the input space as well. However, when inputs are images or texts, more expressive models such as deep neural networks seem more suited than non-parametric methods. In this work, we tackle the question of how to train neural networks to solve structured output prediction tasks, while still benefiting from the versatility and relevance of kernel-induced losses. We design a novel family of deep neural architectures, whose last layer predicts in a data-dependent finite-dimensional subspace of the infinite-dimensional output feature space deriving from the kernel-induced loss. This subspace is chosen as the span of the eigenfunctions of a randomly-approximated version of the empirical kernel covariance operator. Interestingly, this approach unlocks the use of gradient descent algorithms (and consequently of any neural architecture) for structured prediction. Experiments on synthetic tasks as well as real-world supervised graph prediction problems show the relevance of our method. | [
"['Tamim El Ahmad' 'Junjie Yang' 'Pierre Laforgue' \"Florence d'Alché-Buc\"]"
] |
null | null | 2406.09257 | null | null | http://arxiv.org/pdf/2406.09257v1 | 2024-06-13T15:58:37Z | 2024-06-13T15:58:37Z | Assessing Model Generalization in Vicinity | This paper evaluates the generalization ability of classification models on out-of-distribution test sets without depending on ground truth labels. Common approaches often calculate an unsupervised metric related to a specific model property, like confidence or invariance, which correlates with out-of-distribution accuracy. However, these metrics are typically computed for each test sample individually, leading to potential issues caused by spurious model responses, such as overly high or low confidence. To tackle this challenge, we propose incorporating responses from neighboring test samples into the correctness assessment of each individual sample. In essence, if a model consistently demonstrates high correctness scores for nearby samples, it increases the likelihood of correctly predicting the target sample, and vice versa. The resulting scores are then averaged across all test samples to provide a holistic indication of model accuracy. Developed under the vicinal risk formulation, this approach, named vicinal risk proxy (VRP), computes accuracy without relying on labels. We show that applying the VRP method to existing generalization indicators, such as average confidence and effective invariance, consistently improves over these baselines both methodologically and experimentally. This yields a stronger correlation with model accuracy, especially on challenging out-of-distribution test sets. | [
"['Yuchi Liu' 'Yifan Sun' 'Jingdong Wang' 'Liang Zheng']"
] |
null | null | 2406.09262 | null | null | http://arxiv.org/pdf/2406.09262v1 | 2024-06-13T16:02:03Z | 2024-06-13T16:02:03Z | Flexible Heteroscedastic Count Regression with Deep Double Poisson
Networks | Neural networks that can produce accurate, input-conditional uncertainty representations are critical for real-world applications. Recent progress on heteroscedastic continuous regression has shown great promise for calibrated uncertainty quantification on complex tasks, like image regression. However, when these methods are applied to discrete regression tasks, such as crowd counting, ratings prediction, or inventory estimation, they tend to produce predictive distributions with numerous pathologies. We propose to address these issues by training a neural network to output the parameters of a Double Poisson distribution, which we call the Deep Double Poisson Network (DDPN). In contrast to existing methods that are trained to minimize Gaussian negative log likelihood (NLL), DDPNs produce a proper probability mass function over discrete output. Additionally, DDPNs naturally model under-, over-, and equi-dispersion, unlike networks trained with the more rigid Poisson and Negative Binomial parameterizations. We show DDPNs 1) vastly outperform existing discrete models; 2) meet or exceed the accuracy and flexibility of networks trained with Gaussian NLL; 3) produce proper predictive distributions over discrete counts; and 4) exhibit superior out-of-distribution detection. DDPNs can easily be applied to a variety of count regression datasets including tabular, image, point cloud, and text data. | [
"['Spencer Young' 'Porter Jenkins' 'Lonchao Da' 'Jeff Dotson' 'Hua Wei']"
] |
null | null | 2406.09263 | null | null | http://arxiv.org/pdf/2406.09263v2 | 2024-06-14T03:25:22Z | 2024-06-13T16:03:15Z | Generative Inverse Design of Crystal Structures via Diffusion Models
with Transformers | Recent advances in deep learning have enabled the generation of realistic data by training generative models on large datasets of text, images, and audio. While these models have demonstrated exceptional performance in generating novel and plausible data, it remains an open question whether they can effectively accelerate scientific discovery through the data generation and drive significant advancements across various scientific fields. In particular, the discovery of new inorganic materials with promising properties poses a critical challenge, both scientifically and for industrial applications. However, unlike textual or image data, materials, or more specifically crystal structures, consist of multiple types of variables - including lattice vectors, atom positions, and atomic species. This complexity in data give rise to a variety of approaches for representing and generating such data. Consequently, the design choices of generative models for crystal structures remain an open question. In this study, we explore a new type of diffusion model for the generative inverse design of crystal structures, with a backbone based on a Transformer architecture. We demonstrate our models are superior to previous methods in their versatility for generating crystal structures with desired properties. Furthermore, our empirical results suggest that the optimal conditioning methods vary depending on the dataset. | [
"['Izumi Takahara' 'Kiyou Shibata' 'Teruyasu Mizoguchi']"
] |
null | null | 2406.09277 | null | null | http://arxiv.org/pdf/2406.09277v1 | 2024-06-13T16:15:53Z | 2024-06-13T16:15:53Z | End-to-end Streaming model for Low-Latency Speech Anonymization | Speaker anonymization aims to conceal cues to speaker identity while preserving linguistic content. Current machine learning based approaches require substantial computational resources, hindering real-time streaming applications. To address these concerns, we propose a streaming model that achieves speaker anonymization with low latency. The system is trained in an end-to-end autoencoder fashion using a lightweight content encoder that extracts HuBERT-like information, a pretrained speaker encoder that extract speaker identity, and a variance encoder that injects pitch and energy information. These three disentangled representations are fed to a decoder that resynthesizes the speech signal. We present evaluation results from two implementations of our system, a full model that achieves a latency of 230ms, and a lite version (0.1x in size) that further reduces latency to 66ms while maintaining state-of-the-art performance in naturalness, intelligibility, and privacy preservation. | [
"['Waris Quamer' 'Ricardo Gutierrez-Osuna']"
] |
null | null | 2406.09288 | null | null | http://arxiv.org/pdf/2406.09288v1 | 2024-06-13T16:26:37Z | 2024-06-13T16:26:37Z | Zero-Shot Learning Over Large Output Spaces : Utilizing Indirect
Knowledge Extraction from Large Language Models | Extreme Multi-label Learning (XMC) is a task that allocates the most relevant labels for an instance from a predefined label set. Extreme Zero-shot XMC (EZ-XMC) is a special setting of XMC wherein no supervision is provided; only the instances (raw text of the document) and the predetermined label set are given. The scenario is designed to address cold-start problems in categorization and recommendation. Traditional state-of-the-art methods extract pseudo labels from the document title or segments. These labels from the document are used to train a zero-shot bi-encoder model. The main issue with these generated labels is their misalignment with the tagging task. In this work, we propose a framework to train a small bi-encoder model via the feedback from the large language model (LLM), the bi-encoder model encodes the document and labels into embeddings for retrieval. Our approach leverages the zero-shot ability of LLM to assess the correlation between labels and the document instead of using the low-quality labels extracted from the document itself. Our method also guarantees fast inference without the involvement of LLM. The performance of our approach outperforms the SOTA methods on various datasets while retaining a similar training time for large datasets. | [
"['Jinbin Zhang' 'Nasib Ullah' 'Rohit Babbar']"
] |
null | null | 2406.09289 | null | null | http://arxiv.org/pdf/2406.09289v1 | 2024-06-13T16:26:47Z | 2024-06-13T16:26:47Z | Understanding Jailbreak Success: A Study of Latent Space Dynamics in
Large Language Models | Conversational Large Language Models are trained to refuse to answer harmful questions. However, emergent jailbreaking techniques can still elicit unsafe outputs, presenting an ongoing challenge for model alignment. To better understand how different jailbreak types circumvent safeguards, this paper analyses model activations on different jailbreak inputs. We find that it is possible to extract a jailbreak vector from a single class of jailbreaks that works to mitigate jailbreak effectiveness from other classes. This may indicate that different kinds of effective jailbreaks operate via similar internal mechanisms. We investigate a potential common mechanism of harmfulness feature suppression, and provide evidence for its existence by looking at the harmfulness vector component. These findings offer actionable insights for developing more robust jailbreak countermeasures and lay the groundwork for a deeper, mechanistic understanding of jailbreak dynamics in language models. | [
"['Sarah Ball' 'Frauke Kreuter' 'Nina Rimsky']"
] |
null | null | 2406.09291 | null | null | http://arxiv.org/pdf/2406.09291v2 | 2024-07-07T09:32:11Z | 2024-06-13T16:29:06Z | A Flexible, Equivariant Framework for Subgraph GNNs via Graph Products
and Graph Coarsening | Subgraph Graph Neural Networks (Subgraph GNNs) enhance the expressivity of message-passing GNNs by representing graphs as sets of subgraphs. They have shown impressive performance on several tasks, but their complexity limits applications to larger graphs. Previous approaches suggested processing only subsets of subgraphs, selected either randomly or via learnable sampling. However, they make suboptimal subgraph selections or can only cope with very small subset sizes, inevitably incurring performance degradation. This paper introduces a new Subgraph GNNs framework to address these issues. We employ a graph coarsening function to cluster nodes into super-nodes with induced connectivity. The product between the coarsened and the original graph reveals an implicit structure whereby subgraphs are associated with specific sets of nodes. By running generalized message-passing on such graph product, our method effectively implements an efficient, yet powerful Subgraph GNN. Controlling the coarsening function enables meaningful selection of any number of subgraphs while, contrary to previous methods, being fully compatible with standard training techniques. Notably, we discover that the resulting node feature tensor exhibits new, unexplored permutation symmetries. We leverage this structure, characterize the associated linear equivariant layers and incorporate them into the layers of our Subgraph GNN architecture. Extensive experiments on multiple graph learning benchmarks demonstrate that our method is significantly more flexible than previous approaches, as it can seamlessly handle any number of subgraphs, while consistently outperforming baseline approaches. | [
"['Guy Bar-Shalom' 'Yam Eitan' 'Fabrizio Frasca' 'Haggai Maron']"
] |
null | null | 2406.09292 | null | null | http://arxiv.org/pdf/2406.09292v1 | 2024-06-13T16:29:18Z | 2024-06-13T16:29:18Z | Neural Assets: 3D-Aware Multi-Object Scene Synthesis with Image
Diffusion Models | We address the problem of multi-object 3D pose control in image diffusion models. Instead of conditioning on a sequence of text tokens, we propose to use a set of per-object representations, Neural Assets, to control the 3D pose of individual objects in a scene. Neural Assets are obtained by pooling visual representations of objects from a reference image, such as a frame in a video, and are trained to reconstruct the respective objects in a different image, e.g., a later frame in the video. Importantly, we encode object visuals from the reference image while conditioning on object poses from the target frame. This enables learning disentangled appearance and pose features. Combining visual and 3D pose representations in a sequence-of-tokens format allows us to keep the text-to-image architecture of existing models, with Neural Assets in place of text tokens. By fine-tuning a pre-trained text-to-image diffusion model with this information, our approach enables fine-grained 3D pose and placement control of individual objects in a scene. We further demonstrate that Neural Assets can be transferred and recomposed across different scenes. Our model achieves state-of-the-art multi-object editing results on both synthetic 3D scene datasets, as well as two real-world video datasets (Objectron, Waymo Open). | [
"['Ziyi Wu' 'Yulia Rubanova' 'Rishabh Kabra' 'Drew A. Hudson'\n 'Igor Gilitschenski' 'Yusuf Aytar' 'Sjoerd van Steenkiste'\n 'Kelsey R. Allen' 'Thomas Kipf']"
] |
null | null | 2406.09294 | null | null | http://arxiv.org/pdf/2406.09294v1 | 2024-06-13T16:30:03Z | 2024-06-13T16:30:03Z | You Don't Need Data-Augmentation in Self-Supervised Learning | Self-Supervised learning (SSL) with Joint-Embedding Architectures (JEA) has led to outstanding performances. All instantiations of this paradigm were trained using strong and well-established hand-crafted data augmentations, leading to the general belief that they are required for the proper training and performance of such models. On the other hand, generative reconstruction-based models such as BEIT and MAE or Joint-Embedding Predictive Architectures such as I-JEPA have shown strong performance without using data augmentations except masking. In this work, we challenge the importance of invariance and data-augmentation in JEAs at scale. By running a case-study on a recent SSL foundation model - DINOv2 - we show that strong image representations can be obtained with JEAs and only cropping without resizing provided the training data is large enough, reaching state-of-the-art results and using the least amount of augmentation in the literature. Through this study, we also discuss the impact of compute constraints on the outcomes of experimental deep learning research, showing that they can lead to very different conclusions. | [
"['Théo Moutakanni' 'Maxime Oquab' 'Marc Szafraniec' 'Maria Vakalopoulou'\n 'Piotr Bojanowski']"
] |
null | null | 2406.09297 | null | null | http://arxiv.org/pdf/2406.09297v2 | 2024-06-16T03:57:51Z | 2024-06-13T16:33:44Z | MLKV: Multi-Layer Key-Value Heads for Memory Efficient Transformer
Decoding | Auto-regressive inference of transformers benefit greatly from Key-Value (KV) caching, but can lead to major memory bottlenecks as model size, batch size, and sequence length grow at scale. We introduce Multi-Layer Key-Value (MLKV) sharing, a novel approach extending KV sharing across transformer layers to reduce memory usage beyond what was possible with Multi-Query Attention (MQA) and Grouped-Query Attention (GQA). Evaluations on various NLP benchmarks and inference metrics using uptrained Pythia-160M variants demonstrate that MLKV significantly reduces memory usage with minimal performance loss, reducing KV cache size down to a factor of 6x compared to MQA. These results highlight MLKV's potential for efficient deployment of transformer models at scale. We provide code at https://github.com/zaydzuhri/pythia-mlkv | [
"['Zayd Muhammad Kawakibi Zuhri' 'Muhammad Farid Adilazuarda'\n 'Ayu Purwarianti' 'Alham Fikri Aji']"
] |
null | null | 2406.09307 | null | null | http://arxiv.org/pdf/2406.09307v2 | 2024-06-15T11:44:20Z | 2024-06-13T16:41:30Z | A tutorial on fairness in machine learning in healthcare | $textbf{OBJECTIVE}$: Ensuring that machine learning (ML) algorithms are safe and effective within all patient groups, and do not disadvantage particular patients, is essential to clinical decision making and preventing the reinforcement of existing healthcare inequities. The objective of this tutorial is to introduce the medical informatics community to the common notions of fairness within ML, focusing on clinical applications and implementation in practice. $textbf{TARGET AUDIENCE}$: As gaps in fairness arise in a variety of healthcare applications, this tutorial is designed to provide an understanding of fairness, without assuming prior knowledge, to researchers and clinicians who make use of modern clinical data. $textbf{SCOPE}$: We describe the fundamental concepts and methods used to define fairness in ML, including an overview of why models in healthcare may be unfair, a summary and comparison of the metrics used to quantify fairness, and a discussion of some ongoing research. We illustrate some of the fairness methods introduced through a case study of mortality prediction in a publicly available electronic health record dataset. Finally, we provide a user-friendly R package for comprehensive group fairness evaluation, enabling researchers and clinicians to assess fairness in their own ML work. | [
"['Jianhui Gao' 'Benson Chou' 'Zachary R. McCaw' 'Hilary Thurston'\n 'Paul Varghese' 'Chuan Hong' 'Jessica Gronsbell']"
] |
null | null | 2406.09308 | null | null | http://arxiv.org/pdf/2406.09308v1 | 2024-06-13T16:42:06Z | 2024-06-13T16:42:06Z | Transformers meet Neural Algorithmic Reasoners | Transformers have revolutionized machine learning with their simple yet effective architecture. Pre-training Transformers on massive text datasets from the Internet has led to unmatched generalization for natural language understanding (NLU) tasks. However, such language models remain fragile when tasked with algorithmic forms of reasoning, where computations must be precise and robust. To address this limitation, we propose a novel approach that combines the Transformer's language understanding with the robustness of graph neural network (GNN)-based neural algorithmic reasoners (NARs). Such NARs proved effective as generic solvers for algorithmic tasks, when specified in graph form. To make their embeddings accessible to a Transformer, we propose a hybrid architecture with a two-phase training procedure, allowing the tokens in the language model to cross-attend to the node embeddings from the NAR. We evaluate our resulting TransNAR model on CLRS-Text, the text-based version of the CLRS-30 benchmark, and demonstrate significant gains over Transformer-only models for algorithmic reasoning, both in and out of distribution. | [
"['Wilfried Bounsi' 'Borja Ibarz' 'Andrew Dudzik' 'Jessica B. Hamrick'\n 'Larisa Markeeva' 'Alex Vitvitskyi' 'Razvan Pascanu' 'Petar Veličković']"
] |
null | null | 2406.09310 | null | null | http://arxiv.org/pdf/2406.09310v1 | 2024-06-13T16:44:58Z | 2024-06-13T16:44:58Z | Neural networks in non-metric spaces | Leveraging the infinite dimensional neural network architecture we proposed in arXiv:2109.13512v4 and which can process inputs from Fr'echet spaces, and using the universal approximation property shown therein, we now largely extend the scope of this architecture by proving several universal approximation theorems for a vast class of input and output spaces. More precisely, the input space $mathfrak X$ is allowed to be a general topological space satisfying only a mild condition ("quasi-Polish"), and the output space can be either another quasi-Polish space $mathfrak Y$ or a topological vector space $E$. Similarly to arXiv:2109.13512v4, we show furthermore that our neural network architectures can be projected down to "finite dimensional" subspaces with any desirable accuracy, thus obtaining approximating networks that are easy to implement and allow for fast computation and fitting. The resulting neural network architecture is therefore applicable for prediction tasks based on functional data. To the best of our knowledge, this is the first result which deals with such a wide class of input/output spaces and simultaneously guarantees the numerical feasibility of the ensuing architectures. Finally, we prove an obstruction result which indicates that the category of quasi-Polish spaces is in a certain sense the correct category to work with if one aims at constructing approximating architectures on infinite-dimensional spaces $mathfrak X$ which, at the same time, have sufficient expressive power to approximate continuous functions on $mathfrak X$, are specified by a finite number of parameters only and are "stable" with respect to these parameters. | [
"['Luca Galimberti']"
] |
null | null | 2406.09315 | null | null | http://arxiv.org/pdf/2406.09315v1 | 2024-06-13T16:51:33Z | 2024-06-13T16:51:33Z | Vertical LoRA: Dense Expectation-Maximization Interpretation of
Transformers | In this paper, we show how Transformers can be interpreted as dense Expectation-Maximization algorithms performed on Bayesian Nets. Based on the above interpretation, we propose a new model design paradigm, namely Vertical LoRA (VLoRA), which reduces the parameter count dramatically while preserving performance. In VLoRA, a model consists of layers, each of which recursively learns an increment based on the previous layer. We then apply LoRA decomposition to the increments. VLoRA works on the base model, which is orthogonal to LoRA, meaning they can be used together. We do experiments on various tasks and models. The results show that 1) with VLoRA, the Transformer model parameter count can be reduced dramatically and 2) the performance of the original model is preserved. The source code is available at url{https://github.com/neverUseThisName/vlora} | [
"['Zhuolin Fu']"
] |
null | null | 2406.09322 | null | null | http://arxiv.org/pdf/2406.09322v1 | 2024-06-13T17:00:30Z | 2024-06-13T17:00:30Z | Active Inference Meeting Energy-Efficient Control of Parallel and
Identical Machines | We investigate the application of active inference in developing energy-efficient control agents for manufacturing systems. Active inference, rooted in neuroscience, provides a unified probabilistic framework integrating perception, learning, and action, with inherent uncertainty quantification elements. Our study explores deep active inference, an emerging field that combines deep learning with the active inference decision-making framework. Leveraging a deep active inference agent, we focus on controlling parallel and identical machine workstations to enhance energy efficiency. We address challenges posed by the problem's stochastic nature and delayed policy response by introducing tailored enhancements to existing agent architectures. Specifically, we introduce multi-step transition and hybrid horizon methods to mitigate the need for complex planning. Our experimental results demonstrate the effectiveness of these enhancements and highlight the potential of the active inference-based approach. | [
"['Yavar Taheri Yeganeh' 'Mohsen Jafari' 'Andrea Matta']"
] |
null | null | 2406.09329 | null | null | http://arxiv.org/pdf/2406.09329v1 | 2024-06-13T17:07:49Z | 2024-06-13T17:07:49Z | Is Value Learning Really the Main Bottleneck in Offline RL? | While imitation learning requires access to high-quality data, offline reinforcement learning (RL) should, in principle, perform similarly or better with substantially lower data quality by using a value function. However, current results indicate that offline RL often performs worse than imitation learning, and it is often unclear what holds back the performance of offline RL. Motivated by this observation, we aim to understand the bottlenecks in current offline RL algorithms. While poor performance of offline RL is typically attributed to an imperfect value function, we ask: is the main bottleneck of offline RL indeed in learning the value function, or something else? To answer this question, we perform a systematic empirical study of (1) value learning, (2) policy extraction, and (3) policy generalization in offline RL problems, analyzing how these components affect performance. We make two surprising observations. First, we find that the choice of a policy extraction algorithm significantly affects the performance and scalability of offline RL, often more so than the value learning objective. For instance, we show that common value-weighted behavioral cloning objectives (e.g., AWR) do not fully leverage the learned value function, and switching to behavior-constrained policy gradient objectives (e.g., DDPG+BC) often leads to substantial improvements in performance and scalability. Second, we find that a big barrier to improving offline RL performance is often imperfect policy generalization on test-time states out of the support of the training data, rather than policy learning on in-distribution states. We then show that the use of suboptimal but high-coverage data or test-time policy training techniques can address this generalization issue in practice. Specifically, we propose two simple test-time policy improvement methods and show that these methods lead to better performance. | [
"['Seohong Park' 'Kevin Frans' 'Sergey Levine' 'Aviral Kumar']"
] |
null | null | 2406.09335 | null | null | http://arxiv.org/pdf/2406.09335v2 | 2024-06-25T13:47:06Z | 2024-06-13T17:17:31Z | Instance-level quantitative saliency in multiple sclerosis lesion
segmentation | In recent years, explainable methods for artificial intelligence (XAI) have tried to reveal and describe models' decision mechanisms in the case of classification tasks. However, XAI for semantic segmentation and in particular for single instances has been little studied to date. Understanding the process underlying automatic segmentation of single instances is crucial to reveal what information was used to detect and segment a given object of interest. In this study, we proposed two instance-level explanation maps for semantic segmentation based on SmoothGrad and Grad-CAM++ methods. Then, we investigated their relevance for the detection and segmentation of white matter lesions (WML), a magnetic resonance imaging (MRI) biomarker in multiple sclerosis (MS). 687 patients diagnosed with MS for a total of 4043 FLAIR and MPRAGE MRI scans were collected at the University Hospital of Basel, Switzerland. Data were randomly split into training, validation and test sets to train a 3D U-Net for MS lesion segmentation. We observed 3050 true positive (TP), 1818 false positive (FP), and 789 false negative (FN) cases. We generated instance-level explanation maps for semantic segmentation, by developing two XAI methods based on SmoothGrad and Grad-CAM++. We investigated: 1) the distribution of gradients in saliency maps with respect to both input MRI sequences; 2) the model's response in the case of synthetic lesions; 3) the amount of perilesional tissue needed by the model to segment a lesion. Saliency maps (based on SmoothGrad) in FLAIR showed positive values inside a lesion and negative in its neighborhood. Peak values of saliency maps generated for these four groups of volumes presented distributions that differ significantly from one another, suggesting a quantitative nature of the proposed saliency. Contextual information of 7mm around the lesion border was required for their segmentation. | [
"['Federico Spagnolo' 'Nataliia Molchanova' 'Roger Schaer'\n 'Meritxell Bach Cuadra' 'Mario Ocampo Pineda' 'Lester Melie-Garcia'\n 'Cristina Granziera' 'Vincent Andrearczyk' 'Adrien Depeursinge']"
] |
null | null | 2406.09338 | null | null | http://arxiv.org/pdf/2406.09338v1 | 2024-06-13T17:19:43Z | 2024-06-13T17:19:43Z | Learning the Influence Graph of a High-Dimensional Markov Process with
Memory | Motivated by multiple applications in social networks, nervous systems, and financial risk analysis, we consider the problem of learning the underlying (directed) influence graph or causal graph of a high-dimensional multivariate discrete-time Markov process with memory. At any discrete time instant, each observed variable of the multivariate process is a binary string of random length, which is parameterized by an unobservable or hidden [0,1]-valued scalar. The hidden scalars corresponding to the variables evolve according to discrete-time linear stochastic dynamics dictated by the underlying influence graph whose nodes are the variables. We extend an existing algorithm for learning i.i.d. graphical models to this Markovian setting with memory and prove that it can learn the influence graph based on the binary observations using logarithmic (in number of variables or nodes) samples when the degree of the influence graph is bounded. The crucial analytical contribution of this work is the derivation of the sample complexity result by upper and lower bounding the rate of convergence of the observed Markov process with memory to its stationary distribution in terms of the parameters of the influence graph. | [
"['Smita Bagewadi' 'Avhishek Chatterjee']"
] |
null | null | 2406.09346 | null | null | http://arxiv.org/pdf/2406.09346v2 | 2024-06-25T13:25:08Z | 2024-06-13T17:31:02Z | Scoreformer: A Surrogate Model For Large-Scale Prediction of Docking
Scores | In this study, we present ScoreFormer, a novel graph transformer model designed to accurately predict molecular docking scores, thereby optimizing high-throughput virtual screening (HTVS) in drug discovery. The architecture integrates Principal Neighborhood Aggregation (PNA) and Learnable Random Walk Positional Encodings (LRWPE), enhancing the model's ability to understand complex molecular structures and their relationship with their respective docking scores. This approach significantly surpasses traditional HTVS methods and recent Graph Neural Network (GNN) models in both recovery and efficiency due to a wider coverage of the chemical space and enhanced performance. Our results demonstrate that ScoreFormer achieves competitive performance in docking score prediction and offers a substantial 1.65-fold reduction in inference time compared to existing models. We evaluated ScoreFormer across multiple datasets under various conditions, confirming its robustness and reliability in identifying potential drug candidates rapidly. | [
"['Álvaro Ciudad' 'Adrián Morales-Pastor' 'Laura Malo'\n 'Isaac Filella-Mercè' 'Victor Guallar' 'Alexis Molina']"
] |
null | null | 2406.09347 | null | null | http://arxiv.org/pdf/2406.09347v1 | 2024-06-13T17:31:30Z | 2024-06-13T17:31:30Z | Separations in the Representational Capabilities of Transformers and
Recurrent Architectures | Transformer architectures have been widely adopted in foundation models. Due to their high inference costs, there is renewed interest in exploring the potential of efficient recurrent architectures (RNNs). In this paper, we analyze the differences in the representational capabilities of Transformers and RNNs across several tasks of practical relevance, including index lookup, nearest neighbor, recognizing bounded Dyck languages, and string equality. For the tasks considered, our results show separations based on the size of the model required for different architectures. For example, we show that a one-layer Transformer of logarithmic width can perform index lookup, whereas an RNN requires a hidden state of linear size. Conversely, while constant-size RNNs can recognize bounded Dyck languages, we show that one-layer Transformers require a linear size for this task. Furthermore, we show that two-layer Transformers of logarithmic size can perform decision tasks such as string equality or disjointness, whereas both one-layer Transformers and recurrent models require linear size for these tasks. We also show that a log-size two-layer Transformer can implement the nearest neighbor algorithm in its forward pass; on the other hand recurrent models require linear size. Our constructions are based on the existence of $N$ nearly orthogonal vectors in $O(log N)$ dimensional space and our lower bounds are based on reductions from communication complexity problems. We supplement our theoretical results with experiments that highlight the differences in the performance of these architectures on practical-size sequences. | [
"['Satwik Bhattamishra' 'Michael Hahn' 'Phil Blunsom' 'Varun Kanade']"
] |
null | null | 2406.09351 | null | null | http://arxiv.org/pdf/2406.09351v1 | 2024-06-13T17:38:26Z | 2024-06-13T17:38:26Z | On the Expressibility of the Reconstructional Color Refinement | One of the most basic facts related to the famous Ulam reconstruction conjecture is that the connectedness of a graph can be determined by the deck of its vertex-deleted subgraphs, which are considered up to isomorphism. We strengthen this result by proving that connectedness can still be determined when the subgraphs in the deck are given up to equivalence under the color refinement isomorphism test. Consequently, this implies that connectedness is recognizable by Reconstruction Graph Neural Networks, a recently introduced GNN architecture inspired by the reconstruction conjecture (Cotta, Morris, Ribeiro 2021). | [
"['V. Arvind' 'Johannes Köbler' 'Oleg Verbitsky']"
] |
null | null | 2406.09353 | null | null | http://arxiv.org/pdf/2406.09353v1 | 2024-06-13T17:40:15Z | 2024-06-13T17:40:15Z | Enhancing Domain Adaptation through Prompt Gradient Alignment | Prior Unsupervised Domain Adaptation (UDA) methods often aim to train a domain-invariant feature extractor, which may hinder the model from learning sufficiently discriminative features. To tackle this, a line of works based on prompt learning leverages the power of large-scale pre-trained vision-language models to learn both domain-invariant and specific features through a set of domain-agnostic and domain-specific learnable prompts. Those studies typically enforce invariant constraints on representation, output, or prompt space to learn such prompts. Differently, we cast UDA as a multiple-objective optimization problem in which each objective is represented by a domain loss. Under this new framework, we propose aligning per-objective gradients to foster consensus between them. Additionally, to prevent potential overfitting when fine-tuning this deep learning architecture, we penalize the norm of these gradients. To achieve these goals, we devise a practical gradient update procedure that can work under both single-source and multi-source UDA. Empirically, our method consistently surpasses other prompt-based baselines by a large margin on different UDA benchmarks | [
"['Hoang Phan' 'Lam Tran' 'Quyen Tran' 'Trung Le']"
] |
null | null | 2406.09357 | null | null | http://arxiv.org/pdf/2406.09357v1 | 2024-06-13T17:42:57Z | 2024-06-13T17:42:57Z | Advancing Graph Generation through Beta Diffusion | Diffusion models have demonstrated effectiveness in generating natural images and have been extended to generate diverse data types, including graphs. This new generation of diffusion-based graph generative models has demonstrated significant performance improvements over methods that rely on variational autoencoders or generative adversarial networks. It's important to recognize, however, that most of these models employ Gaussian or categorical diffusion processes, which can struggle with sparse and long-tailed data distributions. In our work, we introduce Graph Beta Diffusion (GBD), a diffusion-based generative model particularly adept at capturing diverse graph structures. GBD utilizes a beta diffusion process, tailored for the sparse and range-bounded characteristics of graph adjacency matrices. Furthermore, we have developed a modulation technique that enhances the realism of the generated graphs by stabilizing the generation of critical graph structures, while preserving flexibility elsewhere. The outstanding performance of GBD across three general graph benchmarks and two biochemical graph benchmarks highlights its capability to effectively capture the complexities of real-world graph data. The code will be made available at https://github.com/YH-UtMSB/Graph_Beta_Diffusion | [
"['Yilin He' 'Xinyang Liu' 'Bo Chen' 'Mingyuan Zhou']"
] |
null | null | 2406.09358 | null | null | http://arxiv.org/pdf/2406.09358v1 | 2024-06-13T17:43:41Z | 2024-06-13T17:43:41Z | Understanding Hallucinations in Diffusion Models through Mode
Interpolation | Colloquially speaking, image generation models based upon diffusion processes are frequently said to exhibit "hallucinations," samples that could never occur in the training data. But where do such hallucinations come from? In this paper, we study a particular failure mode in diffusion models, which we term mode interpolation. Specifically, we find that diffusion models smoothly "interpolate" between nearby data modes in the training set, to generate samples that are completely outside the support of the original training distribution; this phenomenon leads diffusion models to generate artifacts that never existed in real data (i.e., hallucinations). We systematically study the reasons for, and the manifestation of this phenomenon. Through experiments on 1D and 2D Gaussians, we show how a discontinuous loss landscape in the diffusion model's decoder leads to a region where any smooth approximation will cause such hallucinations. Through experiments on artificial datasets with various shapes, we show how hallucination leads to the generation of combinations of shapes that never existed. Finally, we show that diffusion models in fact know when they go out of support and hallucinate. This is captured by the high variance in the trajectory of the generated sample towards the final few backward sampling process. Using a simple metric to capture this variance, we can remove over 95% of hallucinations at generation time while retaining 96% of in-support samples. We conclude our exploration by showing the implications of such hallucination (and its removal) on the collapse (and stabilization) of recursive training on synthetic data with experiments on MNIST and 2D Gaussians dataset. We release our code at https://github.com/locuslab/diffusion-model-hallucination. | [
"['Sumukh K Aithal' 'Pratyush Maini' 'Zachary C. Lipton' 'J. Zico Kolter']"
] |
null | null | 2406.09363 | null | null | http://arxiv.org/pdf/2406.09363v2 | 2024-06-19T00:12:35Z | 2024-06-13T17:49:10Z | ElicitationGPT: Text Elicitation Mechanisms via Language Models | Scoring rules evaluate probabilistic forecasts of an unknown state against the realized state and are a fundamental building block in the incentivized elicitation of information and the training of machine learning models. This paper develops mechanisms for scoring elicited text against ground truth text using domain-knowledge-free queries to a large language model (specifically ChatGPT) and empirically evaluates their alignment with human preferences. The empirical evaluation is conducted on peer reviews from a peer-grading dataset and in comparison to manual instructor scores for the peer reviews. | [
"['Yifan Wu' 'Jason Hartline']"
] |
null | null | 2406.09366 | null | null | http://arxiv.org/pdf/2406.09366v1 | 2024-06-13T17:49:56Z | 2024-06-13T17:49:56Z | Towards an Improved Understanding and Utilization of Maximum Manifold
Capacity Representations | Maximum Manifold Capacity Representations (MMCR) is a recent multi-view self-supervised learning (MVSSL) method that matches or surpasses other leading MVSSL methods. MMCR is intriguing because it does not fit neatly into any of the commonplace MVSSL lineages, instead originating from a statistical mechanical perspective on the linear separability of data manifolds. In this paper, we seek to improve our understanding and our utilization of MMCR. To better understand MMCR, we leverage tools from high dimensional probability to demonstrate that MMCR incentivizes alignment and uniformity of learned embeddings. We then leverage tools from information theory to show that such embeddings maximize a well-known lower bound on mutual information between views, thereby connecting the geometric perspective of MMCR to the information-theoretic perspective commonly discussed in MVSSL. To better utilize MMCR, we mathematically predict and experimentally confirm non-monotonic changes in the pretraining loss akin to double descent but with respect to atypical hyperparameters. We also discover compute scaling laws that enable predicting the pretraining loss as a function of gradients steps, batch size, embedding dimension and number of views. We then show that MMCR, originally applied to image data, is performant on multimodal image-text data. By more deeply understanding the theoretical and empirical behavior of MMCR, our work reveals insights on improving MVSSL methods. | [
"['Rylan Schaeffer' 'Victor Lecomte' 'Dhruv Bhandarkar Pai'\n 'Andres Carranza' 'Berivan Isik' 'Alyssa Unell' 'Mikail Khona'\n 'Thomas Yerxa' 'Yann LeCun' 'SueYeon Chung' 'Andrey Gromov'\n 'Ravid Shwartz-Ziv' 'Sanmi Koyejo']"
] |
null | null | 2406.09370 | null | null | http://arxiv.org/pdf/2406.09370v1 | 2024-06-13T17:50:51Z | 2024-06-13T17:50:51Z | Data-dependent and Oracle Bounds on Forgetting in Continual Learning | In continual learning, knowledge must be preserved and re-used between tasks, maintaining good transfer to future tasks and minimizing forgetting of previously learned ones. While several practical algorithms have been devised for this setting, there have been few theoretical works aiming to quantify and bound the degree of Forgetting in general settings. We provide both data-dependent and oracle upper bounds that apply regardless of model and algorithm choice, as well as bounds for Gibbs posteriors. We derive an algorithm inspired by our bounds and demonstrate empirically that our approach yields improved forward and backward transfer. | [
"['Lior Friedman' 'Ron Meir']"
] |
null | null | 2406.09371 | null | null | http://arxiv.org/pdf/2406.09371v1 | 2024-06-13T17:51:00Z | 2024-06-13T17:51:00Z | LRM-Zero: Training Large Reconstruction Models with Synthesized Data | We present LRM-Zero, a Large Reconstruction Model (LRM) trained entirely on synthesized 3D data, achieving high-quality sparse-view 3D reconstruction. The core of LRM-Zero is our procedural 3D dataset, Zeroverse, which is automatically synthesized from simple primitive shapes with random texturing and augmentations (e.g., height fields, boolean differences, and wireframes). Unlike previous 3D datasets (e.g., Objaverse) which are often captured or crafted by humans to approximate real 3D data, Zeroverse completely ignores realistic global semantics but is rich in complex geometric and texture details that are locally similar to or even more intricate than real objects. We demonstrate that our LRM-Zero, trained with our fully synthesized Zeroverse, can achieve high visual quality in the reconstruction of real-world objects, competitive with models trained on Objaverse. We also analyze several critical design choices of Zeroverse that contribute to LRM-Zero's capability and training stability. Our work demonstrates that 3D reconstruction, one of the core tasks in 3D vision, can potentially be addressed without the semantics of real-world objects. The Zeroverse's procedural synthesis code and interactive visualization are available at: https://desaixie.github.io/lrm-zero/. | [
"['Desai Xie' 'Sai Bi' 'Zhixin Shu' 'Kai Zhang' 'Zexiang Xu' 'Yi Zhou'\n 'Sören Pirk' 'Arie Kaufman' 'Xin Sun' 'Hao Tan']"
] |
null | null | 2406.09373 | null | null | http://arxiv.org/pdf/2406.09373v1 | 2024-06-13T17:51:10Z | 2024-06-13T17:51:10Z | Efficient Discrepancy Testing for Learning with Distribution Shift | A fundamental notion of distance between train and test distributions from the field of domain adaptation is discrepancy distance. While in general hard to compute, here we provide the first set of provably efficient algorithms for testing localized discrepancy distance, where discrepancy is computed with respect to a fixed output classifier. These results imply a broad set of new, efficient learning algorithms in the recently introduced model of Testable Learning with Distribution Shift (TDS learning) due to Klivans et al. (2023). Our approach generalizes and improves all prior work on TDS learning: (1) we obtain universal learners that succeed simultaneously for large classes of test distributions, (2) achieve near-optimal error rates, and (3) give exponential improvements for constant depth circuits. Our methods further extend to semi-parametric settings and imply the first positive results for low-dimensional convex sets. Additionally, we separate learning and testing phases and obtain algorithms that run in fully polynomial time at test time. | [
"['Gautam Chandrasekaran' 'Adam R. Klivans' 'Vasilis Kontonis'\n 'Konstantinos Stavropoulos' 'Arsen Vasilyan']"
] |
null | null | 2406.09375 | null | null | http://arxiv.org/pdf/2406.09375v1 | 2024-06-13T17:53:47Z | 2024-06-13T17:53:47Z | Learning conditional distributions on continuous spaces | We investigate sample-based learning of conditional distributions on multi-dimensional unit boxes, allowing for different dimensions of the feature and target spaces. Our approach involves clustering data near varying query points in the feature space to create empirical measures in the target space. We employ two distinct clustering schemes: one based on a fixed-radius ball and the other on nearest neighbors. We establish upper bounds for the convergence rates of both methods and, from these bounds, deduce optimal configurations for the radius and the number of neighbors. We propose to incorporate the nearest neighbors method into neural network training, as our empirical analysis indicates it has better performance in practice. For efficiency, our training process utilizes approximate nearest neighbors search with random binary space partitioning. Additionally, we employ the Sinkhorn algorithm and a sparsity-enforced transport plan. Our empirical findings demonstrate that, with a suitably designed structure, the neural network has the ability to adapt to a suitable level of Lipschitz continuity locally. For reproducibility, our code is available at url{https://github.com/zcheng-a/LCD_kNN}. | [
"['Cyril Bénézet' 'Ziteng Cheng' 'Sebastian Jaimungal']"
] |
null | null | 2406.09384 | null | null | http://arxiv.org/pdf/2406.09384v1 | 2024-06-13T17:57:10Z | 2024-06-13T17:57:10Z | Reflecting on the State of Rehearsal-free Continual Learning with
Pretrained Models | With the advent and recent ubiquity of foundation models, continual learning (CL) has recently shifted from continual training from scratch to the continual adaptation of pretrained models, seeing particular success on rehearsal-free CL benchmarks (RFCL). To achieve this, most proposed methods adapt and restructure parameter-efficient finetuning techniques (PEFT) to suit the continual nature of the problem. Based most often on input-conditional query-mechanisms or regularizations on top of prompt- or adapter-based PEFT, these PEFT-style RFCL (P-RFCL) approaches report peak performances; often convincingly outperforming existing CL techniques. However, on the other end, critical studies have recently highlighted competitive results by training on just the first task or via simple non-parametric baselines. Consequently, questions arise about the relationship between methodological choices in P-RFCL and their reported high benchmark scores. In this work, we tackle these questions to better understand the true drivers behind strong P-RFCL performances, their placement w.r.t. recent first-task adaptation studies, and their relation to preceding CL standards such as EWC or SI. In particular, we show: (1) P-RFCL techniques relying on input-conditional query mechanisms work not because, but rather despite them by collapsing towards standard PEFT shortcut solutions. (2) Indeed, we show how most often, P-RFCL techniques can be matched by a simple and lightweight PEFT baseline. (3) Using this baseline, we identify the implicit bound on tunable parameters when deriving RFCL approaches from PEFT methods as a potential denominator behind P-RFCL efficacy. Finally, we (4) better disentangle continual versus first-task adaptation, and (5) motivate standard RFCL techniques s.a. EWC or SI in light of recent P-RFCL methods. | [
"['Lukas Thede' 'Karsten Roth' 'Olivier J. Hénaff' 'Matthias Bethge'\n 'Zeynep Akata']"
] |
null | null | 2406.09388 | null | null | http://arxiv.org/pdf/2406.09388v1 | 2024-06-13T17:58:39Z | 2024-06-13T17:58:39Z | Exploring the Spectrum of Visio-Linguistic Compositionality and
Recognition | Vision and language models (VLMs) such as CLIP have showcased remarkable zero-shot recognition abilities yet face challenges in visio-linguistic compositionality, particularly in linguistic comprehension and fine-grained image-text alignment. This paper explores the intricate relationship between compositionality and recognition -- two pivotal aspects of VLM capability. We conduct a comprehensive evaluation of existing VLMs, covering both pre-training approaches aimed at recognition and the fine-tuning methods designed to improve compositionality. Our evaluation employs 12 benchmarks for compositionality, along with 21 zero-shot classification and two retrieval benchmarks for recognition. In our analysis from 274 CLIP model checkpoints, we reveal patterns and trade-offs that emerge between compositional understanding and recognition accuracy. Ultimately, this necessitates strategic efforts towards developing models that improve both capabilities, as well as the meticulous formulation of benchmarks for compositionality. We open our evaluation framework at https://github.com/ytaek-oh/vl_compo. | [
"['Youngtaek Oh' 'Pyunghwan Ahn' 'Jinhyung Kim' 'Gwangmo Song'\n 'Soonyoung Lee' 'In So Kweon' 'Junmo Kim']"
] |
null | null | 2406.09390 | null | null | http://arxiv.org/pdf/2406.09390v1 | 2024-06-13T17:59:05Z | 2024-06-13T17:59:05Z | LLAVIDAL: Benchmarking Large Language Vision Models for Daily Activities
of Living | Large Language Vision Models (LLVMs) have demonstrated effectiveness in processing internet videos, yet they struggle with the visually perplexing dynamics present in Activities of Daily Living (ADL) due to limited pertinent datasets and models tailored to relevant cues. To this end, we propose a framework for curating ADL multiview datasets to fine-tune LLVMs, resulting in the creation of ADL-X, comprising 100K RGB video-instruction pairs, language descriptions, 3D skeletons, and action-conditioned object trajectories. We introduce LLAVIDAL, an LLVM capable of incorporating 3D poses and relevant object trajectories to understand the intricate spatiotemporal relationships within ADLs. Furthermore, we present a novel benchmark, ADLMCQ, for quantifying LLVM effectiveness in ADL scenarios. When trained on ADL-X, LLAVIDAL consistently achieves state-of-the-art performance across all ADL evaluation metrics. Qualitative analysis reveals LLAVIDAL's temporal reasoning capabilities in understanding ADL. The link to the dataset is provided at: https://adl-x.github.io/ | [
"['Rajatsubhra Chakraborty' 'Arkaprava Sinha' 'Dominick Reilly'\n 'Manish Kumar Govind' 'Pu Wang' 'Francois Bremond' 'Srijan Das']"
] |
null | null | 2406.09391 | null | null | http://arxiv.org/pdf/2406.09391v1 | 2024-06-13T17:59:06Z | 2024-06-13T17:59:06Z | A More Practical Approach to Machine Unlearning | Machine learning models often incorporate vast amounts of data, raising significant privacy concerns. Machine unlearning, the ability to remove the influence of specific data points from a trained model, addresses these concerns. This paper explores practical methods for implementing machine unlearning, focusing on a first-epoch gradient-ascent approach. Key findings include: 1. Single vs. Multi-Epoch Unlearning: First-epoch gradient unlearning is more effective than multi-epoch gradients. 2. Layer-Based Unlearning: The embedding layer in GPT-2 is crucial for effective unlearning. Gradients from the output layers (11 and 12) have no impact. Efficient unlearning can be achieved using only the embedding layer, halving space complexity. 3. Influence Functions & Scoring: Techniques like Hessian Vector Product and the dot product of activations and tensors are used for quantifying unlearning. 4. Gradient Ascent Considerations: Calibration is necessary to avoid overexposing the model to specific data points during unlearning, which could prematurely terminate the process. 5. Fuzzy Matching vs. Iterative Unlearning: Fuzzy matching techniques shift the model to a new optimum, while iterative unlearning provides a more complete modality. Our empirical evaluation confirms that first-epoch gradient ascent for machine unlearning is more effective than whole-model gradient ascent. These results highlight the potential of machine unlearning for enhancing data privacy and compliance with regulations such as GDPR and CCPA. The study underscores the importance of formal methods to comprehensively evaluate the unlearning process. | [
"['David Zagardo']"
] |
null | null | 2406.09393 | null | null | http://arxiv.org/pdf/2406.09393v1 | 2024-06-13T17:59:09Z | 2024-06-13T17:59:09Z | Improving Autoregressive Training with Dynamic Oracles | Many tasks within NLP can be framed as sequential decision problems, ranging from sequence tagging to text generation. However, for many tasks, the standard training methods, including maximum likelihood (teacher forcing) and scheduled sampling, suffer from exposure bias and a mismatch between metrics employed during training and inference. DAgger provides a solution to mitigate these problems, yet it requires a metric-specific dynamic oracle algorithm, which does not exist for many common metrics like span-based F1, ROUGE, and BLEU. In this paper, we develop these novel dynamic oracles and show they maintain DAgger's no-regret guarantee for decomposable metrics like span-based F1. We evaluate the algorithm's performance on named entity recognition (NER), text summarization, and machine translation (MT). While DAgger with dynamic oracle yields less favorable results in our MT experiments, it outperforms the baseline techniques in NER and text summarization. | [
"['Jianing Yang' 'Harshine Visvanathan' 'Yilin Wang' 'Xinyi Hu'\n 'Matthew Gormley']"
] |
null | null | 2406.09400 | null | null | http://arxiv.org/pdf/2406.09400v1 | 2024-06-13T17:59:29Z | 2024-06-13T17:59:29Z | Yo'LLaVA: Your Personalized Language and Vision Assistant | Large Multimodal Models (LMMs) have shown remarkable capabilities across a variety of tasks (e.g., image captioning, visual question answering). While broad, their knowledge remains generic (e.g., recognizing a dog), and they are unable to handle personalized subjects (e.g., recognizing a user's pet dog). Human reasoning, in contrast, typically operates within the context of specific subjects in our surroundings. For example, one might ask, "What should I buy for my dog's birthday?"; as opposed to a generic inquiry about "What should I buy for a dog's birthday?". Similarly, when looking at a friend's image, the interest lies in seeing their activities (e.g., "my friend is holding a cat"), rather than merely observing generic human actions (e.g., "a man is holding a cat"). In this paper, we introduce the novel task of personalizing LMMs, so that they can have conversations about a specific subject. We propose Yo'LLaVA, which learns to embed a personalized subject into a set of latent tokens given a handful of example images of the subject. Our qualitative and quantitative analyses reveal that Yo'LLaVA can learn the concept more efficiently using fewer tokens and more effectively encode the visual attributes compared to strong prompting baselines (e.g., LLaVA). | [
"['Thao Nguyen' 'Haotian Liu' 'Yuheng Li' 'Mu Cai' 'Utkarsh Ojha'\n 'Yong Jae Lee']"
] |
null | null | 2406.09402 | null | null | http://arxiv.org/pdf/2406.09402v1 | 2024-06-13T17:59:30Z | 2024-06-13T17:59:30Z | Instruct 4D-to-4D: Editing 4D Scenes as Pseudo-3D Scenes Using 2D
Diffusion | This paper proposes Instruct 4D-to-4D that achieves 4D awareness and spatial-temporal consistency for 2D diffusion models to generate high-quality instruction-guided dynamic scene editing results. Traditional applications of 2D diffusion models in dynamic scene editing often result in inconsistency, primarily due to their inherent frame-by-frame editing methodology. Addressing the complexities of extending instruction-guided editing to 4D, our key insight is to treat a 4D scene as a pseudo-3D scene, decoupled into two sub-problems: achieving temporal consistency in video editing and applying these edits to the pseudo-3D scene. Following this, we first enhance the Instruct-Pix2Pix (IP2P) model with an anchor-aware attention module for batch processing and consistent editing. Additionally, we integrate optical flow-guided appearance propagation in a sliding window fashion for more precise frame-to-frame editing and incorporate depth-based projection to manage the extensive data of pseudo-3D scenes, followed by iterative editing to achieve convergence. We extensively evaluate our approach in various scenes and editing instructions, and demonstrate that it achieves spatially and temporally consistent editing results, with significantly enhanced detail and sharpness over the prior art. Notably, Instruct 4D-to-4D is general and applicable to both monocular and challenging multi-camera scenes. Code and more results are available at immortalco.github.io/Instruct-4D-to-4D. | [
"['Linzhan Mou' 'Jun-Kun Chen' 'Yu-Xiong Wang']"
] |
null | null | 2406.09404 | null | null | http://arxiv.org/pdf/2406.09404v1 | 2024-06-13T17:59:32Z | 2024-06-13T17:59:32Z | ConsistDreamer: 3D-Consistent 2D Diffusion for High-Fidelity Scene
Editing | This paper proposes ConsistDreamer - a novel framework that lifts 2D diffusion models with 3D awareness and 3D consistency, thus enabling high-fidelity instruction-guided scene editing. To overcome the fundamental limitation of missing 3D consistency in 2D diffusion models, our key insight is to introduce three synergetic strategies that augment the input of the 2D diffusion model to become 3D-aware and to explicitly enforce 3D consistency during the training process. Specifically, we design surrounding views as context-rich input for the 2D diffusion model, and generate 3D-consistent, structured noise instead of image-independent noise. Moreover, we introduce self-supervised consistency-enforcing training within the per-scene editing procedure. Extensive evaluation shows that our ConsistDreamer achieves state-of-the-art performance for instruction-guided scene editing across various scenes and editing instructions, particularly in complicated large-scale indoor scenes from ScanNet++, with significantly improved sharpness and fine-grained textures. Notably, ConsistDreamer stands as the first work capable of successfully editing complex (e.g., plaid/checkered) patterns. Our project page is at immortalco.github.io/ConsistDreamer. | [
"['Jun-Kun Chen' 'Samuel Rota Bulò' 'Norman Müller' 'Lorenzo Porzi'\n 'Peter Kontschieder' 'Yu-Xiong Wang']"
] |
null | null | 2406.09405 | null | null | http://arxiv.org/pdf/2406.09405v1 | 2024-06-13T17:59:35Z | 2024-06-13T17:59:35Z | Why Warmup the Learning Rate? Underlying Mechanisms and Improvements | It is common in deep learning to warm up the learning rate $eta$, often by a linear schedule between $eta_{text{init}} = 0$ and a predetermined target $eta_{text{trgt}}$. In this paper, we show through systematic experiments using SGD and Adam that the overwhelming benefit of warmup arises from allowing the network to tolerate larger $eta_{text{trgt}}$ by forcing the network to more well-conditioned areas of the loss landscape. The ability to handle larger $eta_{text{trgt}}$ makes hyperparameter tuning more robust while improving the final performance. We uncover different regimes of operation during the warmup period, depending on whether training starts off in a progressive sharpening or sharpness reduction phase, which in turn depends on the initialization and parameterization. Using these insights, we show how $eta_{text{init}}$ can be properly chosen by utilizing the loss catapult mechanism, which saves on the number of warmup steps, in some cases completely eliminating the need for warmup. We also suggest an initialization for the variance in Adam which provides benefits similar to warmup. | [
"['Dayal Singh Kalra' 'Maissam Barkeshli']"
] |
null | null | 2406.09406 | null | null | http://arxiv.org/pdf/2406.09406v2 | 2024-06-14T14:43:26Z | 2024-06-13T17:59:42Z | 4M-21: An Any-to-Any Vision Model for Tens of Tasks and Modalities | Current multimodal and multitask foundation models like 4M or UnifiedIO show promising results, but in practice their out-of-the-box abilities to accept diverse inputs and perform diverse tasks are limited by the (usually rather small) number of modalities and tasks they are trained on. In this paper, we expand upon the capabilities of them by training a single model on tens of highly diverse modalities and by performing co-training on large-scale multimodal datasets and text corpora. This includes training on several semantic and geometric modalities, feature maps from recent state of the art models like DINOv2 and ImageBind, pseudo labels of specialist models like SAM and 4DHumans, and a range of new modalities that allow for novel ways to interact with the model and steer the generation, for example image metadata or color palettes. A crucial step in this process is performing discrete tokenization on various modalities, whether they are image-like, neural network feature maps, vectors, structured data like instance segmentation or human poses, or data that can be represented as text. Through this, we expand on the out-of-the-box capabilities of multimodal models and specifically show the possibility of training one model to solve at least 3x more tasks/modalities than existing ones and doing so without a loss in performance. This enables more fine-grained and controllable multimodal generation capabilities and allows us to study the distillation of models trained on diverse data and objectives into a unified model. We successfully scale the training to a three billion parameter model using tens of modalities and different datasets. The resulting models and training code are open sourced at 4m.epfl.ch. | [
"['Roman Bachmann' 'Oğuzhan Fatih Kar' 'David Mizrahi' 'Ali Garjani'\n 'Mingfei Gao' 'David Griffiths' 'Jiaming Hu' 'Afshin Dehghan'\n 'Amir Zamir']"
] |
null | null | 2406.09408 | null | null | http://arxiv.org/pdf/2406.09408v1 | 2024-06-13T17:59:44Z | 2024-06-13T17:59:44Z | Data Attribution for Text-to-Image Models by Unlearning Synthesized
Images | The goal of data attribution for text-to-image models is to identify the training images that most influence the generation of a new image. We can define "influence" by saying that, for a given output, if a model is retrained from scratch without that output's most influential images, the model should then fail to generate that output image. Unfortunately, directly searching for these influential images is computationally infeasible, since it would require repeatedly retraining from scratch. We propose a new approach that efficiently identifies highly-influential images. Specifically, we simulate unlearning the synthesized image, proposing a method to increase the training loss on the output image, without catastrophic forgetting of other, unrelated concepts. Then, we find training images that are forgotten by proxy, identifying ones with significant loss deviations after the unlearning process, and label these as influential. We evaluate our method with a computationally intensive but "gold-standard" retraining from scratch and demonstrate our method's advantages over previous methods. | [
"['Sheng-Yu Wang' 'Aaron Hertzmann' 'Alexei A. Efros' 'Jun-Yan Zhu'\n 'Richard Zhang']"
] |
null | null | 2406.09412 | null | null | http://arxiv.org/pdf/2406.09412v1 | 2024-06-13T17:59:53Z | 2024-06-13T17:59:53Z | Explore the Limits of Omni-modal Pretraining at Scale | We propose to build omni-modal intelligence, which is capable of understanding any modality and learning universal representations. In specific, we propose a scalable pretraining paradigm, named Multimodal Context (MiCo), which can scale up the numbers of modalities and amount of data, together with the model parameters, in the pretraining process. With MiCo, the pretrained models show significant emergent abilities in multimodal learning, which are evaluated on the following tasks: i) single-modality perception benchmarks of 10 different modalities, ii) 25 cross-modality understanding tasks of retrieval, question-answering, captioning, and iii) 18 multimodal large language model benchmarks. Our models establish 37 new records for state-of-the-art performance. We hope that our research could contribute to the development of omni-modal intelligence. Code and Models are at https://github.com/invictus717/MiCo | [
"['Yiyuan Zhang' 'Handong Li' 'Jing Liu' 'Xiangyu Yue']"
] |
null | null | 2406.09413 | null | null | http://arxiv.org/pdf/2406.09413v1 | 2024-06-13T17:59:56Z | 2024-06-13T17:59:56Z | Interpreting the Weight Space of Customized Diffusion Models | We investigate the space of weights spanned by a large collection of customized diffusion models. We populate this space by creating a dataset of over 60,000 models, each of which is a base model fine-tuned to insert a different person's visual identity. We model the underlying manifold of these weights as a subspace, which we term weights2weights. We demonstrate three immediate applications of this space -- sampling, editing, and inversion. First, as each point in the space corresponds to an identity, sampling a set of weights from it results in a model encoding a novel identity. Next, we find linear directions in this space corresponding to semantic edits of the identity (e.g., adding a beard). These edits persist in appearance across generated samples. Finally, we show that inverting a single image into this space reconstructs a realistic identity, even if the input image is out of distribution (e.g., a painting). Our results indicate that the weight space of fine-tuned diffusion models behaves as an interpretable latent space of identities. | [
"['Amil Dravid' 'Yossi Gandelsman' 'Kuan-Chieh Wang' 'Rameen Abdal'\n 'Gordon Wetzstein' 'Alexei A. Efros' 'Kfir Aberman']"
] |
null | null | 2406.09415 | null | null | http://arxiv.org/pdf/2406.09415v1 | 2024-06-13T17:59:58Z | 2024-06-13T17:59:58Z | An Image is Worth More Than 16x16 Patches: Exploring Transformers on
Individual Pixels | This work does not introduce a new method. Instead, we present an interesting finding that questions the necessity of the inductive bias -- locality in modern computer vision architectures. Concretely, we find that vanilla Transformers can operate by directly treating each individual pixel as a token and achieve highly performant results. This is substantially different from the popular design in Vision Transformer, which maintains the inductive bias from ConvNets towards local neighborhoods (e.g. by treating each 16x16 patch as a token). We mainly showcase the effectiveness of pixels-as-tokens across three well-studied tasks in computer vision: supervised learning for object classification, self-supervised learning via masked autoencoding, and image generation with diffusion models. Although directly operating on individual pixels is less computationally practical, we believe the community must be aware of this surprising piece of knowledge when devising the next generation of neural architectures for computer vision. | [
"['Duy-Kien Nguyen' 'Mahmoud Assran' 'Unnat Jain' 'Martin R. Oswald'\n 'Cees G. M. Snoek' 'Xinlei Chen']"
] |
null | null | 2406.09417 | null | null | http://arxiv.org/pdf/2406.09417v1 | 2024-06-13T17:59:58Z | 2024-06-13T17:59:58Z | Rethinking Score Distillation as a Bridge Between Image Distributions | Score distillation sampling (SDS) has proven to be an important tool, enabling the use of large-scale diffusion priors for tasks operating in data-poor domains. Unfortunately, SDS has a number of characteristic artifacts that limit its usefulness in general-purpose applications. In this paper, we make progress toward understanding the behavior of SDS and its variants by viewing them as solving an optimal-cost transport path from a source distribution to a target distribution. Under this new interpretation, these methods seek to transport corrupted images (source) to the natural image distribution (target). We argue that current methods' characteristic artifacts are caused by (1) linear approximation of the optimal path and (2) poor estimates of the source distribution. We show that calibrating the text conditioning of the source distribution can produce high-quality generation and translation results with little extra overhead. Our method can be easily applied across many domains, matching or beating the performance of specialized methods. We demonstrate its utility in text-to-2D, text-based NeRF optimization, translating paintings to real images, optical illusion generation, and 3D sketch-to-real. We compare our method to existing approaches for score distillation sampling and show that it can produce high-frequency details with realistic colors. | [
"['David McAllister' 'Songwei Ge' 'Jia-Bin Huang' 'David W. Jacobs'\n 'Alexei A. Efros' 'Aleksander Holynski' 'Angjoo Kanazawa']"
] |
null | null | 2406.09441 | null | null | http://arxiv.org/pdf/2406.09441v1 | 2024-06-11T23:14:19Z | 2024-06-11T23:14:19Z | Comment on paper: Position: Rethinking Post-Hoc Search-Based Neural
Approaches for Solving Large-Scale Traveling Salesman Problems | We identify two major issues in the SoftDist paper (Xia et al.): (1) the failure to run all steps of different baselines on the same hardware environment, and (2) the use of inconsistent time measurements when comparing to other baselines. These issues lead to flawed conclusions. When all steps are executed in the same hardware environment, the primary claim made in SoftDist is no longer supported. | [
"['Yimeng Min']"
] |
null | null | 2406.09442 | null | null | http://arxiv.org/pdf/2406.09442v1 | 2024-06-12T00:18:47Z | 2024-06-12T00:18:47Z | An insertable glucose sensor using a compact and cost-effective
phosphorescence lifetime imager and machine learning | Optical continuous glucose monitoring (CGM) systems are emerging for personalized glucose management owing to their lower cost and prolonged durability compared to conventional electrochemical CGMs. Here, we report a computational CGM system, which integrates a biocompatible phosphorescence-based insertable biosensor and a custom-designed phosphorescence lifetime imager (PLI). This compact and cost-effective PLI is designed to capture phosphorescence lifetime images of an insertable sensor through the skin, where the lifetime of the emitted phosphorescence signal is modulated by the local concentration of glucose. Because this phosphorescence signal has a very long lifetime compared to tissue autofluorescence or excitation leakage processes, it completely bypasses these noise sources by measuring the sensor emission over several tens of microseconds after the excitation light is turned off. The lifetime images acquired through the skin are processed by neural network-based models for misalignment-tolerant inference of glucose levels, accurately revealing normal, low (hypoglycemia) and high (hyperglycemia) concentration ranges. Using a 1-mm thick skin phantom mimicking the optical properties of human skin, we performed in vitro testing of the PLI using glucose-spiked samples, yielding 88.8% inference accuracy, also showing resilience to random and unknown misalignments within a lateral distance of ~4.7 mm with respect to the position of the insertable sensor underneath the skin phantom. Furthermore, the PLI accurately identified larger lateral misalignments beyond 5 mm, prompting user intervention for re-alignment. The misalignment-resilient glucose concentration inference capability of this compact and cost-effective phosphorescence lifetime imager makes it an appealing wearable diagnostics tool for real-time tracking of glucose and other biomarkers. | [
"['Artem Goncharov' 'Zoltan Gorocs' 'Ridhi Pradhan' 'Brian Ko'\n 'Ajmal Ajmal' 'Andres Rodriguez' 'David Baum' 'Marcell Veszpremi'\n 'Xilin Yang' 'Maxime Pindrys' 'Tianle Zheng' 'Oliver Wang'\n 'Jessica C. Ramella-Roman' 'Michael J. McShane' 'Aydogan Ozcan']"
] |
null | null | 2406.09443 | null | null | http://arxiv.org/pdf/2406.09443v1 | 2024-06-12T00:53:48Z | 2024-06-12T00:53:48Z | Comparative Analysis of Personalized Voice Activity Detection Systems:
Assessing Real-World Effectiveness | Voice activity detection (VAD) is a critical component in various applications such as speech recognition, speech enhancement, and hands-free communication systems. With the increasing demand for personalized and context-aware technologies, the need for effective personalized VAD systems has become paramount. In this paper, we present a comparative analysis of Personalized Voice Activity Detection (PVAD) systems to assess their real-world effectiveness. We introduce a comprehensive approach to assess PVAD systems, incorporating various performance metrics such as frame-level and utterance-level error rates, detection latency and accuracy, alongside user-level analysis. Through extensive experimentation and evaluation, we provide a thorough understanding of the strengths and limitations of various PVAD variants. This paper advances the understanding of PVAD technology by offering insights into its efficacy and viability in practical applications using a comprehensive set of metrics. | [
"['Satyam Kumar' 'Sai Srujana Buddi' 'Utkarsh Oggy Sarawgi' 'Vineet Garg'\n 'Shivesh Ranjan' 'Ognjen' 'Rudovic' 'Ahmed Hussen Abdelaziz'\n 'Saurabh Adya']"
] |
null | null | 2406.09445 | null | null | http://arxiv.org/pdf/2406.09445v1 | 2024-06-12T01:28:16Z | 2024-06-12T01:28:16Z | A topological analysis of the space of recipes | In recent years, the use of data-driven methods has provided insights into underlying patterns and principles behind culinary recipes. In this exploratory work, we introduce the use of topological data analysis, especially persistent homology, in order to study the space of culinary recipes. In particular, persistent homology analysis provides a set of recipes surrounding the multiscale "holes" in the space of existing recipes. We then propose a method to generate novel ingredient combinations using combinatorial optimization on this topological information. We made biscuits using the novel ingredient combinations, which were confirmed to be acceptable enough by a sensory evaluation study. Our findings indicate that topological data analysis has the potential for providing new tools and insights in the study of culinary recipes. | [
"['Emerson G. Escolar' 'Yuta Shimada' 'Masahiro Yuasa']"
] |
null | null | 2406.09451 | null | null | http://arxiv.org/pdf/2406.09451v1 | 2024-06-12T15:51:00Z | 2024-06-12T15:51:00Z | Simulating Realistic Post-Stroke Reaching Kinematics with Generative
Adversarial Networks | The generalizability of machine learning (ML) models for wearable monitoring in stroke rehabilitation is often constrained by the limited scale and heterogeneity of available data. Data augmentation addresses this challenge by adding computationally derived data to real data to enrich the variability represented in the training set. Traditional augmentation methods, such as rotation, permutation, and time-warping, have shown some benefits in improving classifier performance, but often fail to produce realistic training examples. This study employs Conditional Generative Adversarial Networks (cGANs) to create synthetic kinematic data from a publicly available dataset, closely mimicking the experimentally measured reaching movements of stroke survivors. This approach not only captures the complex temporal dynamics and common movement patterns after stroke, but also significantly enhances the training dataset. By training deep learning models on both synthetic and experimental data, we achieved a substantial enhancement in task classification accuracy: models incorporating synthetic data attained an overall accuracy of 80.2%, significantly higher than the 63.1% seen in models trained solely with real data. These improvements allow for more precise task classification, offering clinicians the potential to monitor patient progress more accurately and tailor rehabilitation interventions more effectively. | [
"['Aaron J. Hadley' 'Christopher L. Pulliam']"
] |
null | null | 2406.09459 | null | null | http://arxiv.org/pdf/2406.09459v1 | 2024-06-12T22:05:51Z | 2024-06-12T22:05:51Z | Ad Auctions for LLMs via Retrieval Augmented Generation | In the field of computational advertising, the integration of ads into the outputs of large language models (LLMs) presents an opportunity to support these services without compromising content integrity. This paper introduces novel auction mechanisms for ad allocation and pricing within the textual outputs of LLMs, leveraging retrieval-augmented generation (RAG). We propose a segment auction where an ad is probabilistically retrieved for each discourse segment (paragraph, section, or entire output) according to its bid and relevance, following the RAG framework, and priced according to competing bids. We show that our auction maximizes logarithmic social welfare, a new notion of welfare that balances allocation efficiency and fairness, and we characterize the associated incentive-compatible pricing rule. These results are extended to multi-ad allocation per segment. An empirical evaluation validates the feasibility and effectiveness of our approach over several ad auction scenarios, and exhibits inherent tradeoffs in metrics as we allow the LLM more flexibility to allocate ads. | [
"['MohammadTaghi Hajiaghayi' 'Sébastien Lahaie' 'Keivan Rezaei' 'Suho Shin']"
] |
null | null | 2406.09463 | null | null | http://arxiv.org/pdf/2406.09463v2 | 2024-06-29T20:27:11Z | 2024-06-13T04:11:01Z | An Effective Software Risk Prediction Management Analysis of Data Using
Machine Learning and Data Mining Method | For one to guarantee higher-quality software development processes, risk management is essential. Furthermore, risks are those that could negatively impact an organization's operations or a project's progress. The appropriate prioritisation of software project risks is a crucial factor in ascertaining the software project's performance features and eventual success. They can be used harmoniously with the same training samples and have good complement and compatibility. We carried out in-depth tests on four benchmark datasets to confirm the efficacy of our CIA approach in closed-world and open-world scenarios, with and without defence. We also present a sequential augmentation parameter optimisation technique that captures the interdependencies of the latest deep learning state-of-the-art WF attack models. To achieve precise software risk assessment, the enhanced crow search algorithm (ECSA) is used to modify the ANFIS settings. Solutions that very slightly alter the local optimum and stay inside it are extracted using the ECSA. ANFIS variable when utilising the ANFIS technique. An experimental validation with NASA 93 dataset and 93 software project values was performed. This method's output presents a clear image of the software risk elements that are essential to achieving project performance. The results of our experiments show that, when compared to other current methods, our integrative fuzzy techniques may perform more accurately and effectively in the evaluation of software project risks. | [
"['Jinxin Xu' 'Yue Wang' 'Ruisi Li' 'Ziyue Wang' 'Qian Zhao']"
] |
null | null | 2406.09465 | null | null | http://arxiv.org/abs/2406.09465v1 | 2024-06-13T04:44:38Z | 2024-06-13T04:44:38Z | Optimal Kernel Orchestration for Tensor Programs with Korch | Kernel orchestration is the task of mapping the computation defined in different operators of a deep neural network (DNN) to the execution of GPU kernels on modern hardware platforms. Prior approaches optimize kernel orchestration by greedily applying operator fusion, which fuses the computation of multiple operators into a single kernel, and miss a variety of optimization opportunities in kernel orchestration. This paper presents Korch, a tensor program optimizer that discovers optimal kernel orchestration strategies for tensor programs. Instead of directly fusing operators, Korch first applies operator fission to decompose tensor operators into a small set of basic tensor algebra primitives. This decomposition enables a diversity of fine-grained, inter-operator optimizations. Next, Korch optimizes kernel orchestration by formalizing it as a constrained optimization problem, leveraging an off-the-shelf binary linear programming solver to discover an optimal orchestration strategy, and generating an executable that can be directly deployed on modern GPU platforms. Evaluation on a variety of DNNs shows that Korch outperforms existing tensor program optimizers by up to 1.7x on V100 GPUs and up to 1.6x on A100 GPUs. Korch is publicly available at https://github.com/humuyan/Korch. | [
"['Muyan Hu' 'Ashwin Venkatram' 'Shreyashri Biswas' 'Balamurugan Marimuthu'\n 'Bohan Hou' 'Gabriele Oliaro' 'Haojie Wang' 'Liyan Zheng' 'Xupeng Miao'\n 'Jidong Zhai']"
] |
null | null | 2406.09474 | null | null | http://arxiv.org/pdf/2406.09474v1 | 2024-06-13T07:46:03Z | 2024-06-13T07:46:03Z | Lightning-Fast Thunderstorm Warnings: Predicting Severe Convective
Environments with Global Neural Weather Models | The recently released suite of AI weather models can produce multi-day, medium-range forecasts within seconds, with a skill on par with state-of-the-art operational forecasts. Traditional AI model evaluation predominantly targets global scores on single levels. Specific prediction tasks, such as severe convective environments, require much more precision on a local scale and with the correct vertical gradients between levels. With a focus on the convective season of global hotspots in 2020, we assess the skill of three top-performing AI models (Pangu-Weather, GraphCast, FourCastNet) for Convective Available Potential Energy (CAPE) and Deep Layer Shear (DLS) at lead-times of up to 10 days against the ERA-5 reanalysis and the IFS operational numerical weather prediction model. Looking at the example of a US tornado outbreak on April 12 and 13, 2020, all models predict elevated CAPE and DLS values multiple days in advance. The spatial structures in the AI models are smoothed in comparison to IFS and ERA-5. The models show differing biases in the prediction of CAPE values, with GraphCast capturing the value distribution the most accurately and FourCastNet showing a consistent underestimation. In seasonal analyses around the globe, we generally see the highest performance by GraphCast and Pangu-Weather, which match or even exceed the performance of IFS. CAPE derived from vertically coarse pressure levels of neural weather models lacks the precision of the vertically fine resolution of numerical models. The promising results here indicate that a direct prediction of CAPE in AI models is likely to be skillful. This would open unprecedented opportunities for fast and inexpensive predictions of severe weather phenomena. By advancing the assessment of AI models towards process-based evaluations we lay the foundation for hazard-driven applications of AI-based weather forecasts. | [
"['Monika Feldmann' 'Tom Beucler' 'Milton Gomez' 'Olivia Martius']"
] |
null | null | 2406.09477 | null | null | http://arxiv.org/pdf/2406.09477v1 | 2024-06-13T09:53:24Z | 2024-06-13T09:53:24Z | Q-S5: Towards Quantized State Space Models | In the quest for next-generation sequence modeling architectures, State Space Models (SSMs) have emerged as a potent alternative to transformers, particularly for their computational efficiency and suitability for dynamical systems. This paper investigates the effect of quantization on the S5 model to understand its impact on model performance and to facilitate its deployment to edge and resource-constrained platforms. Using quantization-aware training (QAT) and post-training quantization (PTQ), we systematically evaluate the quantization sensitivity of SSMs across different tasks like dynamical systems modeling, Sequential MNIST (sMNIST) and most of the Long Range Arena (LRA). We present fully quantized S5 models whose test accuracy drops less than 1% on sMNIST and most of the LRA. We find that performance on most tasks degrades significantly for recurrent weights below 8-bit precision, but that other components can be compressed further without significant loss of performance. Our results further show that PTQ only performs well on language-based LRA tasks whereas all others require QAT. Our investigation provides necessary insights for the continued development of efficient and hardware-optimized SSMs. | [
"['Steven Abreu' 'Jens E. Pedersen' 'Kade M. Heckel' 'Alessandro Pierro']"
] |
null | null | 2406.09481 | null | null | http://arxiv.org/pdf/2406.09481v1 | 2024-06-13T13:00:33Z | 2024-06-13T13:00:33Z | ELF-UA: Efficient Label-Free User Adaptation in Gaze Estimation | We consider the problem of user-adaptive 3D gaze estimation. The performance of person-independent gaze estimation is limited due to interpersonal anatomical differences. Our goal is to provide a personalized gaze estimation model specifically adapted to a target user. Previous work on user-adaptive gaze estimation requires some labeled images of the target person data to fine-tune the model at test time. However, this can be unrealistic in real-world applications, since it is cumbersome for an end-user to provide labeled images. In addition, previous work requires the training data to have both gaze labels and person IDs. This data requirement makes it infeasible to use some of the available data. To tackle these challenges, this paper proposes a new problem called efficient label-free user adaptation in gaze estimation. Our model only needs a few unlabeled images of a target user for the model adaptation. During offline training, we have some labeled source data without person IDs and some unlabeled person-specific data. Our proposed method uses a meta-learning approach to learn how to adapt to a new user with only a few unlabeled images. Our key technical innovation is to use a generalization bound from domain adaptation to define the loss function in meta-learning, so that our method can effectively make use of both the labeled source data and the unlabeled person-specific data during training. Extensive experiments validate the effectiveness of our method on several challenging benchmarks. | [
"['Yong Wu' 'Yang Wang' 'Sanqing Qu' 'Zhijun Li' 'Guang Chen']"
] |
null | null | 2406.09494 | null | null | http://arxiv.org/pdf/2406.09494v1 | 2024-06-13T17:32:32Z | 2024-06-13T17:32:32Z | The Second DISPLACE Challenge : DIarization of SPeaker and LAnguage in
Conversational Environments | The DIarization of SPeaker and LAnguage in Conversational Environments (DISPLACE) 2024 challenge is the second in the series of DISPLACE challenges, which involves tasks of speaker diarization (SD) and language diarization (LD) on a challenging multilingual conversational speech dataset. In the DISPLACE 2024 challenge, we also introduced the task of automatic speech recognition (ASR) on this dataset. The dataset containing 158 hours of speech, consisting of both supervised and unsupervised mono-channel far-field recordings, was released for LD and SD tracks. Further, 12 hours of close-field mono-channel recordings were provided for the ASR track conducted on 5 Indian languages. The details of the dataset, baseline systems and the leader board results are highlighted in this paper. We have also compared our baseline models and the team's performances on evaluation data of DISPLACE-2023 to emphasize the advancements made in this second version of the challenge. | [
"['Shareef Babu Kalluri' 'Prachi Singh' 'Pratik Roy Chowdhuri'\n 'Apoorva Kulkarni' 'Shikha Baghel' 'Pradyoth Hegde' 'Swapnil Sontakke'\n 'Deepak K T' 'S. R. Mahadeva Prasanna' 'Deepu Vijayasenan'\n 'Sriram Ganapathy']"
] |
null | null | 2406.09495 | null | null | http://arxiv.org/pdf/2406.09495v1 | 2024-06-13T17:36:05Z | 2024-06-13T17:36:05Z | Fair Data Generation via Score-based Diffusion Model | The fairness of AI decision-making has garnered increasing attention, leading to the proposal of numerous fairness algorithms. In this paper, we aim not to address this issue by directly introducing fair learning algorithms, but rather by generating entirely new, fair synthetic data from biased datasets for use in any downstream tasks. Additionally, the distribution of test data may differ from that of the training set, potentially impacting the performance of the generated synthetic data in downstream tasks. To address these two challenges, we propose a diffusion model-based framework, FADM: Fairness-Aware Diffusion with Meta-training. FADM introduces two types of gradient induction during the sampling phase of the diffusion model: one to ensure that the generated samples belong to the desired target categories, and another to make the sensitive attributes of the generated samples difficult to classify into any specific sensitive attribute category. To overcome data distribution shifts in the test environment, we train the diffusion model and the two classifiers used for induction within a meta-learning framework. Compared to other baselines, FADM allows for flexible control over the categories of the generated samples and exhibits superior generalization capability. Experiments on real datasets demonstrate that FADM achieves better accuracy and optimal fairness in downstream tasks. | [
"['Yujie Lin' 'Dong Li' 'Chen Zhao' 'Minglai Shao']"
] |
null | null | 2406.09509 | null | null | http://arxiv.org/pdf/2406.09509v1 | 2024-06-13T18:00:24Z | 2024-06-13T18:00:24Z | CleanDiffuser: An Easy-to-use Modularized Library for Diffusion Models
in Decision Making | Leveraging the powerful generative capability of diffusion models (DMs) to build decision-making agents has achieved extensive success. However, there is still a demand for an easy-to-use and modularized open-source library that offers customized and efficient development for DM-based decision-making algorithms. In this work, we introduce CleanDiffuser, the first DM library specifically designed for decision-making algorithms. By revisiting the roles of DMs in the decision-making domain, we identify a set of essential sub-modules that constitute the core of CleanDiffuser, allowing for the implementation of various DM algorithms with simple and flexible building blocks. To demonstrate the reliability and flexibility of CleanDiffuser, we conduct comprehensive evaluations of various DM algorithms implemented with CleanDiffuser across an extensive range of tasks. The analytical experiments provide a wealth of valuable design choices and insights, reveal opportunities and challenges, and lay a solid groundwork for future research. CleanDiffuser will provide long-term support to the decision-making community, enhancing reproducibility and fostering the development of more robust solutions. The code and documentation of CleanDiffuser are open-sourced on the https://github.com/CleanDiffuserTeam/CleanDiffuser. | [
"['Zibin Dong' 'Yifu Yuan' 'Jianye Hao' 'Fei Ni' 'Yi Ma' 'Pengyi Li'\n 'Yan Zheng']"
] |
null | null | 2406.09513 | null | null | http://arxiv.org/pdf/2406.09513v1 | 2024-06-13T18:07:04Z | 2024-06-13T18:07:04Z | Fair GLASSO: Estimating Fair Graphical Models with Unbiased Statistical
Behavior | We propose estimating Gaussian graphical models (GGMs) that are fair with respect to sensitive nodal attributes. Many real-world models exhibit unfair discriminatory behavior due to biases in data. Such discrimination is known to be exacerbated when data is equipped with pairwise relationships encoded in a graph. Additionally, the effect of biased data on graphical models is largely underexplored. We thus introduce fairness for graphical models in the form of two bias metrics to promote balance in statistical similarities across nodal groups with different sensitive attributes. Leveraging these metrics, we present Fair GLASSO, a regularized graphical lasso approach to obtain sparse Gaussian precision matrices with unbiased statistical dependencies across groups. We also propose an efficient proximal gradient algorithm to obtain the estimates. Theoretically, we express the tradeoff between fair and accurate estimated precision matrices. Critically, this includes demonstrating when accuracy can be preserved in the presence of a fairness regularizer. On top of this, we study the complexity of Fair GLASSO and demonstrate that our algorithm enjoys a fast convergence rate. Our empirical validation includes synthetic and real-world simulations that illustrate the value and effectiveness of our proposed optimization problem and iterative algorithm. | [
"['Madeline Navarro' 'Samuel Rey' 'Andrei Buciulea' 'Antonio G. Marques'\n 'Santiago Segarra']"
] |
null | null | 2406.09520 | null | null | http://arxiv.org/abs/2406.09520v1 | 2024-06-13T18:16:27Z | 2024-06-13T18:16:27Z | A Systematic Review of Generative AI for Teaching and Learning Practice | The use of generative artificial intelligence (GenAI) in academia is a subjective and hotly debated topic. Currently, there are no agreed guidelines towards the usage of GenAI systems in higher education (HE) and, thus, it is still unclear how to make effective use of the technology for teaching and learning practice. This paper provides an overview of the current state of research on GenAI for teaching and learning in HE. To this end, this study conducted a systematic review of relevant studies indexed by Scopus, using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. The search criteria revealed a total of 625 research papers, of which 355 met the final inclusion criteria. The findings from the review showed the current state and the future trends in documents, citations, document sources/authors, keywords, and co-authorship. The research gaps identified suggest that while some authors have looked at understanding the detection of AI-generated text, it may be beneficial to understand how GenAI can be incorporated into supporting the educational curriculum for assessments, teaching, and learning delivery. Furthermore, there is a need for additional interdisciplinary, multidimensional studies in HE through collaboration. This will strengthen the awareness and understanding of students, tutors, and other stakeholders, which will be instrumental in formulating guidelines, frameworks, and policies for GenAI usage. | [
"['Bayode Ogunleye' 'Kudirat Ibilola Zakariyyah' 'Oluwaseun Ajao'\n 'Olakunle Olayinka' 'Hemlata Sharma']"
] |
null | null | 2406.09529 | null | null | http://arxiv.org/pdf/2406.09529v1 | 2024-06-13T18:37:24Z | 2024-06-13T18:37:24Z | Differentiable Reasoning about Knowledge Graphs with Region-based Graph
Neural Networks | Methods for knowledge graph (KG) completion need to capture semantic regularities and use these regularities to infer plausible knowledge that is not explicitly stated. Most embedding-based methods are opaque in the kinds of regularities they can capture, although region-based KG embedding models have emerged as a more transparent alternative. By modeling relations as geometric regions in high-dimensional vector spaces, such models can explicitly capture semantic regularities in terms of the spatial arrangement of these regions. Unfortunately, existing region-based approaches are severely limited in the kinds of rules they can capture. We argue that this limitation arises because the considered regions are defined as the Cartesian product of two-dimensional regions. As an alternative, in this paper, we propose RESHUFFLE, a simple model based on ordering constraints that can faithfully capture a much larger class of rule bases than existing approaches. Moreover, the embeddings in our framework can be learned by a monotonic Graph Neural Network (GNN), which effectively acts as a differentiable rule base. This approach has the important advantage that embeddings can be easily updated as new knowledge is added to the KG. At the same time, since the resulting representations can be used similarly to standard KG embeddings, our approach is significantly more efficient than existing approaches to differentiable reasoning. | [
"['Aleksandar Pavlovic' 'Emanuel Sallinger' 'Steven Schockaert']"
] |
null | null | 2406.09534 | null | null | http://arxiv.org/pdf/2406.09534v1 | 2024-06-13T18:44:48Z | 2024-06-13T18:44:48Z | FeatNavigator: Automatic Feature Augmentation on Tabular Data | Data-centric AI focuses on understanding and utilizing high-quality, relevant data in training machine learning (ML) models, thereby increasing the likelihood of producing accurate and useful results. Automatic feature augmentation, aiming to augment the initial base table with useful features from other tables, is critical in data preparation as it improves model performance, robustness, and generalizability. While recent works have investigated automatic feature augmentation, most of them have limited capabilities in utilizing all useful features as many of them are in candidate tables not directly joinable with the base table. Worse yet, with numerous join paths leading to these distant features, existing solutions fail to fully exploit them within a reasonable compute budget. We present FeatNavigator, an effective and efficient framework that explores and integrates high-quality features in relational tables for ML models. FeatNavigator evaluates a feature from two aspects: (1) the intrinsic value of a feature towards an ML task (i.e., feature importance) and (2) the efficacy of a join path connecting the feature to the base table (i.e., integration quality). FeatNavigator strategically selects a small set of available features and their corresponding join paths to train a feature importance estimation model and an integration quality prediction model. Furthermore, FeatNavigator's search algorithm exploits both estimated feature importance and integration quality to identify the optimized feature augmentation plan. Our experimental results show that FeatNavigator outperforms state-of-the-art solutions on five public datasets by up to 40.1% in ML model performance. | [
"['Jiaming Liang' 'Chuan Lei' 'Xiao Qin' 'Jiani Zhang'\n 'Asterios Katsifodimos' 'Christos Faloutsos' 'Huzefa Rangwala']"
] |
null | null | 2406.09535 | null | null | http://arxiv.org/abs/2406.09535v1 | 2024-06-13T18:47:52Z | 2024-06-13T18:47:52Z | CircuitVAE: Efficient and Scalable Latent Circuit Optimization | Automatically designing fast and space-efficient digital circuits is challenging because circuits are discrete, must exactly implement the desired logic, and are costly to simulate. We address these challenges with CircuitVAE, a search algorithm that embeds computation graphs in a continuous space and optimizes a learned surrogate of physical simulation by gradient descent. By carefully controlling overfitting of the simulation surrogate and ensuring diverse exploration, our algorithm is highly sample-efficient, yet gracefully scales to large problem instances and high sample budgets. We test CircuitVAE by designing binary adders across a large range of sizes, IO timing constraints, and sample budgets. Our method excels at designing large circuits, where other algorithms struggle: compared to reinforcement learning and genetic algorithms, CircuitVAE typically finds 64-bit adders which are smaller and faster using less than half the sample budget. We also find CircuitVAE can design state-of-the-art adders in a real-world chip, demonstrating that our method can outperform commercial tools in a realistic setting. | [
"['Jialin Song' 'Aidan Swope' 'Robert Kirby' 'Rajarshi Roy' 'Saad Godil'\n 'Jonathan Raiman' 'Bryan Catanzaro']"
] |
null | null | 2406.09547 | null | null | http://arxiv.org/abs/2406.09547v2 | 2024-06-18T11:49:36Z | 2024-06-13T19:28:08Z | FLea: Addressing Data Scarcity and Label Skew in Federated Learning via
Privacy-preserving Feature Augmentation | Federated Learning (FL) enables model development by leveraging data distributed across numerous edge devices without transferring local data to a central server. However, existing FL methods still face challenges when dealing with scarce and label-skewed data across devices, resulting in local model overfitting and drift, consequently hindering the performance of the global model. In response to these challenges, we propose a pioneering framework called FLea, incorporating the following key components: i) A global feature buffer that stores activation-target pairs shared from multiple clients to support local training. This design mitigates local model drift caused by the absence of certain classes; ii) A feature augmentation approach based on local and global activation mix-ups for local training. This strategy enlarges the training samples, thereby reducing the risk of local overfitting; iii) An obfuscation method to minimize the correlation between intermediate activations and the source data, enhancing the privacy of shared features. To verify the superiority of FLea, we conduct extensive experiments using a wide range of data modalities, simulating different levels of local data scarcity and label skew. The results demonstrate that FLea consistently outperforms state-of-the-art FL counterparts (among 13 of the experimented 18 settings, the improvement is over 5% while concurrently mitigating the privacy vulnerabilities associated with shared features. Code is available at https://github.com/XTxiatong/FLea.git. | [
"['Tong Xia' 'Abhirup Ghosh' 'Xinchi Qiu' 'Cecilia Mascolo']"
] |
null | null | 2406.09548 | null | null | http://arxiv.org/pdf/2406.09548v1 | 2024-06-13T19:29:37Z | 2024-06-13T19:29:37Z | Between Randomness and Arbitrariness: Some Lessons for Reliable Machine
Learning at Scale | To develop rigorous knowledge about ML models -- and the systems in which they are embedded -- we need reliable measurements. But reliable measurement is fundamentally challenging, and touches on issues of reproducibility, scalability, uncertainty quantification, epistemology, and more. This dissertation addresses criteria needed to take reliability seriously: both criteria for designing meaningful metrics, and for methodologies that ensure that we can dependably and efficiently measure these metrics at scale and in practice. In doing so, this dissertation articulates a research vision for a new field of scholarship at the intersection of machine learning, law, and policy. Within this frame, we cover topics that fit under three different themes: (1) quantifying and mitigating sources of arbitrariness in ML, (2) taming randomness in uncertainty estimation and optimization algorithms, in order to achieve scalability without sacrificing reliability, and (3) providing methods for evaluating generative-AI systems, with specific focuses on quantifying memorization in language models and training latent diffusion models on open-licensed data. By making contributions in these three themes, this dissertation serves as an empirical proof by example that research on reliable measurement for machine learning is intimately and inescapably bound up with research in law and policy. These different disciplines pose similar research questions about reliable measurement in machine learning. They are, in fact, two complementary sides of the same research vision, which, broadly construed, aims to construct machine-learning systems that cohere with broader societal values. | [
"['A. Feder Cooper']"
] |
null | null | 2406.09549 | null | null | http://arxiv.org/pdf/2406.09549v1 | 2024-06-13T19:30:32Z | 2024-06-13T19:30:32Z | Exploring Syntactic Patterns in Urdu: A Deep Dive into Dependency
Analysis | Parsing is the process of breaking a sentence into its grammatical components and identifying the syntactic structure of the sentence. The syntactically correct sentence structure is achieved by assigning grammatical labels to its constituents using lexicon and syntactic rules. In linguistics, parser is extremely useful due to the number of different applications like name entity recognition, QA systems and information extraction, etc. The two most common techniques used for parsing are phrase structure and dependency Structure. Because Urdu is a low-resource language, there has been little progress in building an Urdu parser. A comparison of several parsers revealed that the dependency parsing approach is better suited for order-free languages such as Urdu. We have made significant progress in parsing Urdu, a South Asian language with a complex morphology. For Urdu dependency parsing, a basic feature model consisting of word location, word head, and dependency relation is employed as a starting point, followed by more complex feature models. The dependency tagset is designed after careful consideration of the complex morphological structure of the Urdu language, word order variation, and lexical ambiguity and it contains 22 tags. Our dataset comprises of sentences from news articles, and we tried to include sentences of different complexity (which is quite challenging), to get reliable results. All experiments are performed using MaltParser, exploring all 9 algorithms and classifiers. We have achieved a 70 percent overall best-labeled accuracy (LA), as well as an 84 percent overall best-unlabeled attachment score (UAS) using the Nivreeager algorithm. The comparison of output data with treebank test data that has been manually parsed is then used to carry out error assessment and to identify the errors produced by the parser. | [
"['Nudrat Habib']"
] |
null | null | 2406.09551 | null | null | http://arxiv.org/pdf/2406.09551v1 | 2024-06-13T19:35:58Z | 2024-06-13T19:35:58Z | Embedding machine-learnt sub-grid variability improves climate model
biases | The under-representation of cloud formation is a long-standing bias associated with climate simulations. Parameterisation schemes are required to capture cloud processes within current climate models but have known biases. We overcome these biases by embedding a Multi-Output Gaussian Process (MOGP) trained on high resolution Unified Model simulations to represent the variability of temperature and specific humidity within a climate model. A trained MOGP model is coupled in-situ with a simplified Atmospheric General Circulation Model named SPEEDY. The temperature and specific humidity profiles of SPEEDY are perturbed at fixed intervals according to the variability predicted from the MOGP. Ten-year predictions are generated for both control and ML-hybrid models. The hybrid model reduces the global precipitation bias by 18% and over the tropics by 22%. To further understand the drivers of these improvements, physical quantities of interest are explored, such as the distribution of lifted index values and the alteration of the Hadley cell. The control and hybrid set-ups are also run in a plus 4K sea-surface temperature experiment to explore the effects of the approach on patterns relating to cloud cover and precipitation in a warmed climate setting. | [
"['Daniel Giles' 'James Briant' 'Cyril J. Morcrette' 'Serge Guillas']"
] |
null | null | 2406.09556 | null | null | http://arxiv.org/pdf/2406.09556v2 | 2024-06-18T14:12:18Z | 2024-06-13T19:43:38Z | $S^3$ -- Semantic Signal Separation | Topic models are useful tools for discovering latent semantic structures in large textual corpora. Topic modeling historically relied on bag-of-words representations of language. This approach makes models sensitive to the presence of stop words and noise, and does not utilize potentially useful contextual information. Recent efforts have been oriented at incorporating contextual neural representations in topic modeling and have been shown to outperform classical topic models. These approaches are, however, typically slow, volatile and still require preprocessing for optimal results. We present Semantic Signal Separation ($S^3$), a theory-driven topic modeling approach in neural embedding spaces. $S^3$ conceptualizes topics as independent axes of semantic space, and uncovers these with blind-source separation. Our approach provides the most diverse, highly coherent topics, requires no preprocessing, and is demonstrated to be the fastest contextually sensitive topic model to date. We offer an implementation of $S^3$, among other approaches, in the Turftopic Python package. | [
"['Márton Kardos' 'Jan Kostkan' 'Arnault-Quentin Vermillet'\n 'Kristoffer Nielbo' 'Kenneth Enevoldsen' 'Roberta Rocca']"
] |
null | null | 2406.09559 | null | null | http://arxiv.org/pdf/2406.09559v1 | 2024-06-13T19:55:20Z | 2024-06-13T19:55:20Z | Decoding the Diversity: A Review of the Indic AI Research Landscape | This review paper provides a comprehensive overview of large language model (LLM) research directions within Indic languages. Indic languages are those spoken in the Indian subcontinent, including India, Pakistan, Bangladesh, Sri Lanka, Nepal, and Bhutan, among others. These languages have a rich cultural and linguistic heritage and are spoken by over 1.5 billion people worldwide. With the tremendous market potential and growing demand for natural language processing (NLP) based applications in diverse languages, generative applications for Indic languages pose unique challenges and opportunities for research. Our paper deep dives into the recent advancements in Indic generative modeling, contributing with a taxonomy of research directions, tabulating 84 recent publications. Research directions surveyed in this paper include LLM development, fine-tuning existing LLMs, development of corpora, benchmarking and evaluation, as well as publications around specific techniques, tools, and applications. We found that researchers across the publications emphasize the challenges associated with limited data availability, lack of standardization, and the peculiar linguistic complexities of Indic languages. This work aims to serve as a valuable resource for researchers and practitioners working in the field of NLP, particularly those focused on Indic languages, and contributes to the development of more accurate and efficient LLM applications for these languages. | [
"['Sankalp KJ' 'Vinija Jain' 'Sreyoshi Bhaduri' 'Tamoghna Roy'\n 'Aman Chadha']"
] |
null | null | 2406.09561 | null | null | http://arxiv.org/pdf/2406.09561v1 | 2024-06-13T20:00:06Z | 2024-06-13T20:00:06Z | Label Noise Robustness for Domain-Agnostic Fair Corrections via Nearest
Neighbors Label Spreading | Last-layer retraining methods have emerged as an efficient framework for correcting existing base models. Within this framework, several methods have been proposed to deal with correcting models for subgroup fairness with and without group membership information. Importantly, prior work has demonstrated that many methods are susceptible to noisy labels. To this end, we propose a drop-in correction for label noise in last-layer retraining, and demonstrate that it achieves state-of-the-art worst-group accuracy for a broad range of symmetric label noise and across a wide variety of datasets exhibiting spurious correlations. Our proposed approach uses label spreading on a latent nearest neighbors graph and has minimal computational overhead compared to existing methods. | [
"['Nathan Stromberg' 'Rohan Ayyagari' 'Sanmi Koyejo' 'Richard Nock'\n 'Lalitha Sankar']"
] |
null | null | 2406.09563 | null | null | http://arxiv.org/pdf/2406.09563v1 | 2024-06-13T20:12:09Z | 2024-06-13T20:12:09Z | e-COP : Episodic Constrained Optimization of Policies | In this paper, we present the $texttt{e-COP}$ algorithm, the first policy optimization algorithm for constrained Reinforcement Learning (RL) in episodic (finite horizon) settings. Such formulations are applicable when there are separate sets of optimization criteria and constraints on a system's behavior. We approach this problem by first establishing a policy difference lemma for the episodic setting, which provides the theoretical foundation for the algorithm. Then, we propose to combine a set of established and novel solution ideas to yield the $texttt{e-COP}$ algorithm that is easy to implement and numerically stable, and provide a theoretical guarantee on optimality under certain scaling assumptions. Through extensive empirical analysis using benchmarks in the Safety Gym suite, we show that our algorithm has similar or better performance than SoTA (non-episodic) algorithms adapted for the episodic setting. The scalability of the algorithm opens the door to its application in safety-constrained Reinforcement Learning from Human Feedback for Large Language or Diffusion Models. | [
"['Akhil Agnihotri' 'Rahul Jain' 'Deepak Ramachandran' 'Sahil Singla']"
] |
null | null | 2406.09564 | null | null | http://arxiv.org/pdf/2406.09564v1 | 2024-06-13T20:12:46Z | 2024-06-13T20:12:46Z | Towards Domain Adaptive Neural Contextual Bandits | Contextual bandit algorithms are essential for solving real-world decision making problems. In practice, collecting a contextual bandit's feedback from different domains may involve different costs. For example, measuring drug reaction from mice (as a source domain) and humans (as a target domain). Unfortunately, adapting a contextual bandit algorithm from a source domain to a target domain with distribution shift still remains a major challenge and largely unexplored. In this paper, we introduce the first general domain adaptation method for contextual bandits. Our approach learns a bandit model for the target domain by collecting feedback from the source domain. Our theoretical analysis shows that our algorithm maintains a sub-linear regret bound even adapting across domains. Empirical results show that our approach outperforms the state-of-the-art contextual bandit algorithms on real-world datasets. | [
"['Ziyan Wang' 'Hao Wang']"
] |
null | null | 2406.09567 | null | null | http://arxiv.org/pdf/2406.09567v1 | 2024-06-13T20:18:16Z | 2024-06-13T20:18:16Z | Causal Fine-Tuning and Effect Calibration of Non-Causal Predictive
Models | This paper proposes techniques to enhance the performance of non-causal models for causal inference using data from randomized experiments. In domains like advertising, customer retention, and precision medicine, non-causal models that predict outcomes under no intervention are often used to score individuals and rank them according to the expected effectiveness of an intervention (e.g, an ad, a retention incentive, a nudge). However, these scores may not perfectly correspond to intervention effects due to the inherent non-causal nature of the models. To address this limitation, we propose causal fine-tuning and effect calibration, two techniques that leverage experimental data to refine the output of non-causal models for different causal tasks, including effect estimation, effect ordering, and effect classification. They are underpinned by two key advantages. First, they can effectively integrate the predictive capabilities of general non-causal models with the requirements of a causal task in a specific context, allowing decision makers to support diverse causal applications with a "foundational" scoring model. Second, through simulations and an empirical example, we demonstrate that they can outperform the alternative of building a causal-effect model from scratch, particularly when the available experimental data is limited and the non-causal scores already capture substantial information about the relative sizes of causal effects. Overall, this research underscores the practical advantages of combining experimental data with non-causal models to support causal applications. | [
"['Carlos Fernández-Loría' 'Yanfang Hou' 'Foster Provost' 'Jennifer Hill']"
] |
null | null | 2406.09570 | null | null | http://arxiv.org/pdf/2406.09570v1 | 2024-06-13T20:22:38Z | 2024-06-13T20:22:38Z | Improving Consistency Models with Generator-Induced Coupling | Consistency models are promising generative models as they distill the multi-step sampling of score-based diffusion in a single forward pass of a neural network. Without access to sampling trajectories of a pre-trained diffusion model, consistency training relies on proxy trajectories built on an independent coupling between the noise and data distributions. Refining this coupling is a key area of improvement to make it more adapted to the task and reduce the resulting randomness in the training process. In this work, we introduce a novel coupling associating the input noisy data with their generated output from the consistency model itself, as a proxy to the inaccessible diffusion flow output. Our affordable approach exploits the inherent capacity of consistency models to compute the transport map in a single step. We provide intuition and empirical evidence of the relevance of our generator-induced coupling (GC), which brings consistency training closer to score distillation. Consequently, our method not only accelerates consistency training convergence by significant amounts but also enhances the resulting performance. The code is available at: https://github.com/thibautissenhuth/consistency_GC. | [
"['Thibaut Issenhuth' 'Ludovic Dos Santos' 'Jean-Yves Franceschi'\n 'Alain Rakotomamonjy']"
] |
null | null | 2406.09574 | null | null | http://arxiv.org/pdf/2406.09574v1 | 2024-06-13T20:25:52Z | 2024-06-13T20:25:52Z | Online Bandit Learning with Offline Preference Data | Reinforcement Learning with Human Feedback (RLHF) is at the core of fine-tuning methods for generative AI models for language and images. Such feedback is often sought as rank or preference feedback from human raters, as opposed to eliciting scores since the latter tends to be very noisy. On the other hand, RL theory and algorithms predominantly assume that a reward feedback is available. In particular, approaches for online learning that can be helpful in adaptive data collection via active learning cannot incorporate offline preference data. In this paper, we adopt a finite-armed linear bandit model as a prototypical model of online learning. We consider an offline preference dataset to be available generated by an expert of unknown 'competence'. We propose $texttt{warmPref-PS}$, a posterior sampling algorithm for online learning that can be warm-started with an offline dataset with noisy preference feedback. We show that by modeling the competence of the expert that generated it, we are able to use such a dataset most effectively. We support our claims with novel theoretical analysis of its Bayesian regret, as well as extensive empirical evaluation of an approximate algorithm which performs substantially better (almost 25 to 50% regret reduction in our studies) as compared to baselines. | [
"['Akhil Agnihotri' 'Rahul Jain' 'Deepak Ramachandran' 'Zheng Wen']"
] |
null | null | 2406.09581 | null | null | http://arxiv.org/pdf/2406.09581v1 | 2024-06-13T20:39:59Z | 2024-06-13T20:39:59Z | A Review of 315 Benchmark and Test Functions for Machine Learning
Optimization Algorithms and Metaheuristics with Mathematical and Visual
Descriptions | In the rapidly evolving optimization and metaheuristics domains, the efficacy of algorithms is crucially determined by the benchmark (test) functions. While several functions have been developed and derived over the past decades, little information is available on the mathematical and visual description, range of suitability, and applications of many such functions. To bridge this knowledge gap, this review provides an exhaustive survey of more than 300 benchmark functions used in the evaluation of optimization and metaheuristics algorithms. This review first catalogs benchmark and test functions based on their characteristics, complexity, properties, visuals, and domain implications to offer a wide view that aids in selecting appropriate benchmarks for various algorithmic challenges. This review also lists the 25 most commonly used functions in the open literature and proposes two new, highly dimensional, dynamic and challenging functions that could be used for testing new algorithms. Finally, this review identifies gaps in current benchmarking practices and suggests directions for future research. | [
"['M. Z. Naser' 'Mohammad Khaled al-Bashiti' 'Arash Teymori Gharah Tapeh'\n 'Armin Dadras Eslamlou' 'Ahmed Naser' 'Venkatesh Kodur' 'Rami Hawileeh'\n 'Jamal Abdalla' 'Nima Khodadadi' 'Amir H. Gandomi']"
] |
null | null | 2406.09588 | null | null | http://arxiv.org/pdf/2406.09588v1 | 2024-06-13T21:02:03Z | 2024-06-13T21:02:03Z | Color Equivariant Network | Group equivariant convolutional neural networks have been designed for a variety of geometric transformations from 2D and 3D rotation groups, to semi-groups such as scale. Despite the improved interpretability, accuracy and generalizability afforded by these architectures, group equivariant networks have seen limited application in the context of perceptual quantities such as hue and saturation, even though their variation can lead to significant reductions in classification performance. In this paper, we introduce convolutional neural networks equivariant to variations in hue and saturation by design. To achieve this, we leverage the observation that hue and saturation transformations can be identified with the 2D rotation and 1D translation groups respectively. Our hue-, saturation-, and fully color-equivariant networks achieve equivariance to these perceptual transformations without an increase in network parameters. We demonstrate the utility of our networks on synthetic and real world datasets where color and lighting variations are commonplace. | [
"[\"Felix O'Mahony\" 'Yulong Yang' 'Christine Allen-Blanchette']"
] |
null | null | 2406.09592 | null | null | http://arxiv.org/pdf/2406.09592v1 | 2024-06-13T21:17:25Z | 2024-06-13T21:17:25Z | On Value Iteration Convergence in Connected MDPs | This paper establishes that an MDP with a unique optimal policy and ergodic associated transition matrix ensures the convergence of various versions of the Value Iteration algorithm at a geometric rate that exceeds the discount factor {gamma} for both discounted and average-reward criteria. | [
"['Arsenii Mustafin' 'Alex Olshevsky' 'Ioannis Ch. Paschalidis']"
] |
null | null | 2406.09606 | null | null | http://arxiv.org/pdf/2406.09606v2 | 2024-06-27T22:06:19Z | 2024-06-13T22:34:58Z | Cross-Modality Program Representation Learning for Electronic Design
Automation with High-Level Synthesis | In recent years, domain-specific accelerators (DSAs) have gained popularity for applications such as deep learning and autonomous driving. To facilitate DSA designs, programmers use high-level synthesis (HLS) to compile a high-level description written in C/C++ into a design with low-level hardware description languages that eventually synthesize DSAs on circuits. However, creating a high-quality HLS design still demands significant domain knowledge, particularly in microarchitecture decisions expressed as textit{pragmas}. Thus, it is desirable to automate such decisions with the help of machine learning for predicting the quality of HLS designs, requiring a deeper understanding of the program that consists of original code and pragmas. Naturally, these programs can be considered as sequence data. In addition, these programs can be compiled and converted into a control data flow graph (CDFG). But existing works either fail to leverage both modalities or combine the two in shallow or coarse ways. We propose ProgSG, a model that allows interaction between the source code sequence modality and the graph modality in a deep and fine-grained way. To alleviate the scarcity of labeled designs, a pre-training method is proposed based on a suite of compiler's data flow analysis tasks. Experimental results show that ProgSG reduces the RMSE of design performance predictions by up to $22%$, and identifies designs with an average of $1.10times$ and $1.26times$ (up to $8.17times$ and $13.31times$) performance improvement in design space exploration (DSE) task compared to HARP and AutoDSE, respectively. | [
"['Zongyue Qin' 'Yunsheng Bai' 'Atefeh Sohrabizadeh' 'Zijian Ding'\n 'Ziniu Hu' 'Yizhou Sun' 'Jason Cong']"
] |
null | null | 2406.09612 | null | null | http://arxiv.org/pdf/2406.09612v1 | 2024-06-13T22:44:08Z | 2024-06-13T22:44:08Z | Automated Molecular Concept Generation and Labeling with Large Language
Models | Artificial intelligence (AI) is significantly transforming scientific research. Explainable AI methods, such as concept-based models (CMs), are promising for driving new scientific discoveries because they make predictions based on meaningful concepts and offer insights into the prediction process. In molecular science, however, explainable CMs are not as common compared to black-box models like Graph Neural Networks (GNNs), primarily due to their requirement for predefined concepts and manual label for each instance, which demand domain knowledge and can be labor-intensive. This paper introduces a novel framework for Automated Molecular Concept (AutoMolCo) generation and labeling. AutoMolCo leverages the knowledge in Large Language Models (LLMs) to automatically generate predictive molecular concepts and label them for each molecule. Such procedures are repeated through iterative interactions with LLMs to refine concepts, enabling simple linear models on the refined concepts to outperform GNNs and LLM in-context learning on several benchmarks. The whole AutoMolCo framework is automated without any human knowledge inputs in either concept generation, labeling, or refinement, thereby surpassing the limitations of extant CMs while maintaining their explainability and allowing easy intervention. Through systematic experiments on MoleculeNet and High-Throughput Experimentation (HTE) datasets, we demonstrate that the AutoMolCo-induced explainable CMs are beneficial and promising for molecular science research. | [
"['Shichang Zhang' 'Botao Xia' 'Zimin Zhang' 'Qianli Wu' 'Fang Sun'\n 'Ziniu Hu' 'Yizhou Sun']"
] |
null | null | 2406.09614 | null | null | http://arxiv.org/pdf/2406.09614v1 | 2024-06-13T22:45:13Z | 2024-06-13T22:45:13Z | Trainability issues in quantum policy gradients | This research explores the trainability of Parameterized Quantum circuit-based policies in Reinforcement Learning, an area that has recently seen a surge in empirical exploration. While some studies suggest improved sample complexity using quantum gradient estimation, the efficient trainability of these policies remains an open question. Our findings reveal significant challenges, including standard Barren Plateaus with exponentially small gradients and gradient explosion. These phenomena depend on the type of basis-state partitioning and mapping these partitions onto actions. For a polynomial number of actions, a trainable window can be ensured with a polynomial number of measurements if a contiguous-like partitioning of basis-states is employed. These results are empirically validated in a multi-armed bandit environment. | [
"['André Sequeira' 'Luis Paulo Santos' 'Luis Soares Barbosa']"
] |
null | null | 2406.09624 | null | null | http://arxiv.org/pdf/2406.09624v1 | 2024-06-13T23:19:48Z | 2024-06-13T23:19:48Z | DrivAerNet++: A Large-Scale Multimodal Car Dataset with Computational
Fluid Dynamics Simulations and Deep Learning Benchmarks | We present DrivAerNet++, the largest and most comprehensive multimodal dataset for aerodynamic car design. DrivAerNet++ comprises 8,000 diverse car designs modeled with high-fidelity computational fluid dynamics (CFD) simulations. The dataset includes diverse car configurations such as fastback, notchback, and estateback, with different underbody and wheel designs to represent both internal combustion engines and electric vehicles. Each entry in the dataset features detailed 3D meshes, parametric models, aerodynamic coefficients, and extensive flow and surface field data, along with segmented parts for car classification and point cloud data. This dataset supports a wide array of machine learning applications including data-driven design optimization, generative modeling, surrogate model training, CFD simulation acceleration, and geometric classification. With more than 39 TB of publicly available engineering data, DrivAerNet++ fills a significant gap in available resources, providing high-quality, diverse data to enhance model training, promote generalization, and accelerate automotive design processes. Along with rigorous dataset validation, we also provide ML benchmarking results on the task of aerodynamic drag prediction, showcasing the breadth of applications supported by our dataset. This dataset is set to significantly impact automotive design and broader engineering disciplines by fostering innovation and improving the fidelity of aerodynamic evaluations. | [
"['Mohamed Elrefaie' 'Florin Morar' 'Angela Dai' 'Faez Ahmed']"
] |
null | null | 2406.09630 | null | null | http://arxiv.org/pdf/2406.09630v1 | 2024-06-13T23:40:34Z | 2024-06-13T23:40:34Z | Muharaf: Manuscripts of Handwritten Arabic Dataset for Cursive Text
Recognition | We present the Manuscripts of Handwritten Arabic~(Muharaf) dataset, which is a machine learning dataset consisting of more than 1,600 historic handwritten page images transcribed by experts in archival Arabic. Each document image is accompanied by spatial polygonal coordinates of its text lines as well as basic page elements. This dataset was compiled to advance the state of the art in handwritten text recognition (HTR), not only for Arabic manuscripts but also for cursive text in general. The Muharaf dataset includes diverse handwriting styles and a wide range of document types, including personal letters, diaries, notes, poems, church records, and legal correspondences. In this paper, we describe the data acquisition pipeline, notable dataset features, and statistics. We also provide a preliminary baseline result achieved by training convolutional neural networks using this data. | [
"['Mehreen Saeed' 'Adrian Chan' 'Anupam Mijar' 'Joseph Moukarzel'\n 'Georges Habchi' 'Carlos Younes' 'Amin Elias' 'Chau-Wai Wong'\n 'Akram Khater']"
] |
null | null | 2406.09638 | null | null | http://arxiv.org/pdf/2406.09638v1 | 2024-06-14T00:07:52Z | 2024-06-14T00:07:52Z | RASPNet: A Benchmark Dataset for Radar Adaptive Signal Processing
Applications | This work presents a large-scale dataset for radar adaptive signal processing (RASP) applications, aimed at supporting the development of data-driven models within the radar community. The dataset, called RASPNet, consists of 100 realistic scenarios compiled over a variety of topographies and land types from across the contiguous United States, designed to reflect a diverse array of real-world environments. Within each scenario, RASPNet consists of 10,000 clutter realizations from an airborne radar setting, which can be utilized for radar algorithm development and evaluation. RASPNet intends to fill a prominent gap in the availability of a large-scale, realistic dataset that standardizes the evaluation of adaptive radar processing techniques. We describe its construction, organization, and several potential applications, which includes a transfer learning example to demonstrate how RASPNet can be leveraged for realistic adaptive radar processing scenarios. | [
"['Shyam Venkatasubramanian' 'Bosung Kang' 'Ali Pezeshki'\n 'Muralidhar Rangaswamy' 'Vahid Tarokh']"
] |
null | null | 2406.09639 | null | null | http://arxiv.org/pdf/2406.09639v1 | 2024-06-14T00:08:04Z | 2024-06-14T00:08:04Z | TGB 2.0: A Benchmark for Learning on Temporal Knowledge Graphs and
Heterogeneous Graphs | Multi-relational temporal graphs are powerful tools for modeling real-world data, capturing the evolving and interconnected nature of entities over time. Recently, many novel models are proposed for ML on such graphs intensifying the need for robust evaluation and standardized benchmark datasets. However, the availability of such resources remains scarce and evaluation faces added complexity due to reproducibility issues in experimental protocols. To address these challenges, we introduce Temporal Graph Benchmark 2.0 (TGB 2.0), a novel benchmarking framework tailored for evaluating methods for predicting future links on Temporal Knowledge Graphs and Temporal Heterogeneous Graphs with a focus on large-scale datasets, extending the Temporal Graph Benchmark. TGB 2.0 facilitates comprehensive evaluations by presenting eight novel datasets spanning five domains with up to 53 million edges. TGB 2.0 datasets are significantly larger than existing datasets in terms of number of nodes, edges, or timestamps. In addition, TGB 2.0 provides a reproducible and realistic evaluation pipeline for multi-relational temporal graphs. Through extensive experimentation, we observe that 1) leveraging edge-type information is crucial to obtain high performance, 2) simple heuristic baselines are often competitive with more complex methods, 3) most methods fail to run on our largest datasets, highlighting the need for research on more scalable methods. | [
"['Julia Gastinger' 'Shenyang Huang' 'Mikhail Galkin' 'Erfan Loghmani'\n 'Ali Parviz' 'Farimah Poursafaei' 'Jacob Danovitch' 'Emanuele Rossi'\n 'Ioannis Koutis' 'Heiner Stuckenschmidt' 'Reihaneh Rabbany'\n 'Guillaume Rabusseau']"
] |
null | null | 2406.09643 | null | null | http://arxiv.org/pdf/2406.09643v1 | 2024-06-14T00:24:29Z | 2024-06-14T00:24:29Z | Reinforced Decoder: Towards Training Recurrent Neural Networks for Time
Series Forecasting | Recurrent neural network-based sequence-to-sequence models have been extensively applied for multi-step-ahead time series forecasting. These models typically involve a decoder trained using either its previous forecasts or the actual observed values as the decoder inputs. However, relying on self-generated predictions can lead to the rapid accumulation of errors over multiple steps, while using the actual observations introduces exposure bias as these values are unavailable during the extrapolation stage. In this regard, this study proposes a novel training approach called reinforced decoder, which introduces auxiliary models to generate alternative decoder inputs that remain accessible when extrapolating. Additionally, a reinforcement learning algorithm is utilized to dynamically select the optimal inputs to improve accuracy. Comprehensive experiments demonstrate that our approach outperforms representative training methods over several datasets. Furthermore, the proposed approach also exhibits promising performance when generalized to self-attention-based sequence-to-sequence forecasting models. | [
"['Qi Sima' 'Xinze Zhang' 'Yukun Bao' 'Siyue Yang' 'Liang Shen']"
] |
null | null | 2406.09648 | null | null | http://arxiv.org/pdf/2406.09648v1 | 2024-06-14T00:40:31Z | 2024-06-14T00:40:31Z | An Intrinsic Vector Heat Network | Vector fields are widely used to represent and model flows for many science and engineering applications. This paper introduces a novel neural network architecture for learning tangent vector fields that are intrinsically defined on manifold surfaces embedded in 3D. Previous approaches to learning vector fields on surfaces treat vectors as multi-dimensional scalar fields, using traditional scalar-valued architectures to process channels individually, thus fail to preserve fundamental intrinsic properties of the vector field. The core idea of this work is to introduce a trainable vector heat diffusion module to spatially propagate vector-valued feature data across the surface, which we incorporate into our proposed architecture that consists of vector-valued neurons. Our architecture is invariant to rigid motion of the input, isometric deformation, and choice of local tangent bases, and is robust to discretizations of the surface. We evaluate our Vector Heat Network on triangle meshes, and empirically validate its invariant properties. We also demonstrate the effectiveness of our method on the useful industrial application of quadrilateral mesh generation. | [
"['Alexander Gao' 'Maurice Chu' 'Mubbasir Kapadia' 'Ming C. Lin'\n 'Hsueh-Ti Derek Liu']"
] |
null | null | 2406.09654 | null | null | http://arxiv.org/pdf/2406.09654v1 | 2024-06-14T01:24:01Z | 2024-06-14T01:24:01Z | Coralai: Intrinsic Evolution of Embodied Neural Cellular Automata
Ecosystems | This paper presents Coralai, a framework for exploring diverse ecosystems of Neural Cellular Automata (NCA). Organisms in Coralai utilize modular, GPU-accelerated Taichi kernels to interact, enact environmental changes, and evolve through local survival, merging, and mutation operations implemented with HyperNEAT and PyTorch. We provide an exploratory experiment implementing physics inspired by slime mold behavior showcasing the emergence of competition between sessile and mobile organisms, cycles of resource depletion and recovery, and symbiosis between diverse organisms. We conclude by outlining future work to discover simulation parameters through measures of multi-scale complexity and diversity. Code for Coralai is available at https://github.com/aidanbx/coralai , video demos are available at https://www.youtube.com/watch?v=NL8IZQY02-8 . | [
"['Aidan Barbieux' 'Rodrigo Canaan']"
] |
null | null | 2406.09656 | null | null | http://arxiv.org/pdf/2406.09656v1 | 2024-06-14T01:36:52Z | 2024-06-14T01:36:52Z | RSEND: Retinex-based Squeeze and Excitation Network with Dark Region
Detection for Efficient Low Light Image Enhancement | Images captured under low-light scenarios often suffer from low quality. Previous CNN-based deep learning methods often involve using Retinex theory. Nevertheless, most of them cannot perform well in more complicated datasets like LOL-v2 while consuming too much computational resources. Besides, some of these methods require sophisticated training at different stages, making the procedure even more time-consuming and tedious. In this paper, we propose a more accurate, concise, and one-stage Retinex theory based framework, RSEND. RSEND first divides the low-light image into the illumination map and reflectance map, then captures the important details in the illumination map and performs light enhancement. After this step, it refines the enhanced gray-scale image and does element-wise matrix multiplication with the reflectance map. By denoising the output it has from the previous step, it obtains the final result. In all the steps, RSEND utilizes Squeeze and Excitation network to better capture the details. Comprehensive quantitative and qualitative experiments show that our Efficient Retinex model significantly outperforms other CNN-based models, achieving a PSNR improvement ranging from 0.44 dB to 4.2 dB in different datasets and even outperforms transformer-based models in the LOL-v2-real dataset. | [
"['Jingcheng Li' 'Ye Qiao' 'Haocheng Xu' 'Sitao Huang']"
] |
null | null | 2406.09657 | null | null | http://arxiv.org/pdf/2406.09657v1 | 2024-06-14T02:04:59Z | 2024-06-14T02:04:59Z | ScaLES: Scalable Latent Exploration Score for Pre-Trained Generative
Networks | We develop Scalable Latent Exploration Score (ScaLES) to mitigate over-exploration in Latent Space Optimization (LSO), a popular method for solving black-box discrete optimization problems. LSO utilizes continuous optimization within the latent space of a Variational Autoencoder (VAE) and is known to be susceptible to over-exploration, which manifests in unrealistic solutions that reduce its practicality. ScaLES is an exact and theoretically motivated method leveraging the trained decoder's approximation of the data distribution. ScaLES can be calculated with any existing decoder, e.g. from a VAE, without additional training, architectural changes, or access to the training data. Our evaluation across five LSO benchmark tasks and three VAE architectures demonstrates that ScaLES enhances the quality of the solutions while maintaining high objective values, leading to improvements over existing solutions. We believe that new avenues to LSO will be opened by ScaLES ability to identify out of distribution areas, differentiability, and computational tractability. Open source code for ScaLES is available at https://github.com/OmerRonen/scales. | [
"['Omer Ronen' 'Ahmed Imtiaz Humayun' 'Randall Balestriero'\n 'Richard Baraniuk' 'Bin Yu']"
] |
null | null | 2406.09675 | null | null | http://arxiv.org/pdf/2406.09675v1 | 2024-06-14T02:56:57Z | 2024-06-14T02:56:57Z | Benchmarking Spectral Graph Neural Networks: A Comprehensive Study on
Effectiveness and Efficiency | With the recent advancements in graph neural networks (GNNs), spectral GNNs have received increasing popularity by virtue of their specialty in capturing graph signals in the frequency domain, demonstrating promising capability in specific tasks. However, few systematic studies have been conducted on assessing their spectral characteristics. This emerging family of models also varies in terms of designs and settings, leading to difficulties in comparing their performance and deciding on the suitable model for specific scenarios, especially for large-scale tasks. In this work, we extensively benchmark spectral GNNs with a focus on the frequency perspective. We analyze and categorize over 30 GNNs with 27 corresponding filters. Then, we implement these spectral models under a unified framework with dedicated graph computations and efficient training schemes. Thorough experiments are conducted on the spectral models with inclusive metrics on effectiveness and efficiency, offering practical guidelines on evaluating and selecting spectral GNNs with desirable performance. Our implementation enables application on larger graphs with comparable performance and less overhead, which is available at: https://github.com/gdmnl/Spectral-GNN-Benchmark. | [
"['Ningyi Liao' 'Haoyu Liu' 'Zulun Zhu' 'Siqiang Luo'\n 'Laks V. S. Lakshmanan']"
] |
null | null | 2406.09680 | null | null | http://arxiv.org/pdf/2406.09680v1 | 2024-06-14T03:05:05Z | 2024-06-14T03:05:05Z | Heterogeneous Federated Learning with Convolutional and Spiking Neural
Networks | Federated learning (FL) has emerged as a promising paradigm for training models on decentralized data while safeguarding data privacy. Most existing FL systems, however, assume that all machine learning models are of the same type, although it becomes more likely that different edge devices adopt different types of AI models, including both conventional analogue artificial neural networks (ANNs) and biologically more plausible spiking neural networks (SNNs). This diversity empowers the efficient handling of specific tasks and requirements, showcasing the adaptability and versatility of edge computing platforms. One main challenge of such heterogeneous FL system lies in effectively aggregating models from the local devices in a privacy-preserving manner. To address the above issue, this work benchmarks FL systems containing both convoluntional neural networks (CNNs) and SNNs by comparing various aggregation approaches, including federated CNNs, federated SNNs, federated CNNs for SNNs, federated SNNs for CNNs, and federated CNNs with SNN fusion. Experimental results demonstrate that the CNN-SNN fusion framework exhibits the best performance among the above settings on the MNIST dataset. Additionally, intriguing phenomena of competitive suppression are noted during the convergence process of multi-model FL. | [
"['Yingchao Yu' 'Yuping Yan' 'Jisong Cai' 'Yaochu Jin']"
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.