categories
string | doi
string | id
string | year
float64 | venue
string | link
string | updated
string | published
string | title
string | abstract
string | authors
list |
|---|---|---|---|---|---|---|---|---|---|---|
null | null | 2405.04393 | null | null | http://arxiv.org/pdf/2405.04393v1 | 2024-05-07T15:14:51Z | 2024-05-07T15:14:51Z | Efficient Online Set-valued Classification with Bandit Feedback | Conformal prediction is a distribution-free method that wraps a given machine learning model and returns a set of plausible labels that contain the true label with a prescribed coverage rate. In practice, the empirical coverage achieved highly relies on fully observed label information from data both in the training phase for model fitting and the calibration phase for quantile estimation. This dependency poses a challenge in the context of online learning with bandit feedback, where a learner only has access to the correctness of actions (i.e., pulled an arm) but not the full information of the true label. In particular, when the pulled arm is incorrect, the learner only knows that the pulled one is not the true class label, but does not know which label is true. Additionally, bandit feedback further results in a smaller labeled dataset for calibration, limited to instances with correct actions, thereby affecting the accuracy of quantile estimation. To address these limitations, we propose Bandit Class-specific Conformal Prediction (BCCP), offering coverage guarantees on a class-specific granularity. Using an unbiased estimation of an estimand involving the true label, BCCP trains the model and makes set-valued inferences through stochastic gradient descent. Our approach overcomes the challenges of sparsely labeled data in each iteration and generalizes the reliability and applicability of conformal prediction to online decision-making environments. | [
"['Zhou Wang' 'Xingye Qiao']"
]
|
null | null | 2405.04396 | null | null | http://arxiv.org/pdf/2405.04396v1 | 2024-05-07T15:18:21Z | 2024-05-07T15:18:21Z | Predicting Transonic Flowfields in Non-Homogeneous Unstructured Grids
Using Autoencoder Graph Convolutional Networks | This paper focuses on addressing challenges posed by non-homogeneous unstructured grids, commonly used in Computational Fluid Dynamics (CFD). Their prevalence in CFD scenarios has motivated the exploration of innovative approaches for generating reduced-order models. The core of our approach centers on geometric deep learning, specifically the utilization of graph convolutional network (GCN). The novel Autoencoder GCN architecture enhances prediction accuracy by propagating information to distant nodes and emphasizing influential points. This architecture, with GCN layers and encoding/decoding modules, reduces dimensionality based on pressure-gradient values. The autoencoder structure improves the network capability to identify key features, contributing to a more robust and accurate predictive model. To validate the proposed methodology, we analyzed two different test cases: wing-only model and wing--body configuration. Precise reconstruction of steady-state distributed quantities within a two-dimensional parametric space underscores the reliability and versatility of the implemented approach. | [
"['Gabriele Immordino' 'Andrea Vaiuso' 'Andrea Da Ronch' 'Marcello Righi']"
]
|
null | null | 2405.04404 | null | null | http://arxiv.org/pdf/2405.04404v1 | 2024-05-07T15:30:14Z | 2024-05-07T15:30:14Z | Vision Mamba: A Comprehensive Survey and Taxonomy | State Space Model (SSM) is a mathematical model used to describe and analyze the behavior of dynamic systems. This model has witnessed numerous applications in several fields, including control theory, signal processing, economics and machine learning. In the field of deep learning, state space models are used to process sequence data, such as time series analysis, natural language processing (NLP) and video understanding. By mapping sequence data to state space, long-term dependencies in the data can be better captured. In particular, modern SSMs have shown strong representational capabilities in NLP, especially in long sequence modeling, while maintaining linear time complexity. Notably, based on the latest state-space models, Mamba merges time-varying parameters into SSMs and formulates a hardware-aware algorithm for efficient training and inference. Given its impressive efficiency and strong long-range dependency modeling capability, Mamba is expected to become a new AI architecture that may outperform Transformer. Recently, a number of works have attempted to study the potential of Mamba in various fields, such as general vision, multi-modal, medical image analysis and remote sensing image analysis, by extending Mamba from natural language domain to visual domain. To fully understand Mamba in the visual domain, we conduct a comprehensive survey and present a taxonomy study. This survey focuses on Mamba's application to a variety of visual tasks and data types, and discusses its predecessors, recent advances and far-reaching impact on a wide range of domains. Since Mamba is now on an upward trend, please actively notice us if you have new findings, and new progress on Mamba will be included in this survey in a timely manner and updated on the Mamba project at https://github.com/lx6c78/Vision-Mamba-A-Comprehensive-Survey-and-Taxonomy. | [
"['Xiao Liu' 'Chenxu Zhang' 'Lei Zhang']"
]
|
null | null | 2405.04405 | null | null | http://arxiv.org/pdf/2405.04405v2 | 2024-05-09T08:51:37Z | 2024-05-07T15:31:58Z | Weakly-Supervised Residual Evidential Learning for Multi-Instance
Uncertainty Estimation | Uncertainty estimation (UE), as an effective means of quantifying predictive uncertainty, is crucial for safe and reliable decision-making, especially in high-risk scenarios. Existing UE schemes usually assume that there are completely-labeled samples to support fully-supervised learning. In practice, however, many UE tasks often have no sufficiently-labeled data to use, such as the Multiple Instance Learning (MIL) with only weak instance annotations. To bridge this gap, this paper, for the first time, addresses the weakly-supervised issue of Multi-Instance UE (MIUE) and proposes a new baseline scheme, Multi-Instance Residual Evidential Learning (MIREL). Particularly, at the fine-grained instance UE with only weak supervision, we derive a multi-instance residual operator through the Fundamental Theorem of Symmetric Functions. On this operator derivation, we further propose MIREL to jointly model the high-order predictive distribution at bag and instance levels for MIUE. Extensive experiments empirically demonstrate that our MIREL not only could often make existing MIL networks perform better in MIUE, but also could surpass representative UE methods by large margins, especially in instance-level UE tasks. Our source code is available at https://github.com/liupei101/MIREL. | [
"['Pei Liu' 'Luping Ji']"
]
|
null | null | 2405.04407 | null | null | http://arxiv.org/pdf/2405.04407v2 | 2024-05-17T12:15:25Z | 2024-05-07T15:35:30Z | Super-Exponential Regret for UCT, AlphaGo and Variants | We improve the proofs of the lower bounds of Coquelin and Munos (2007) that demonstrate that UCT can have $exp(dotsexp(1)dots)$ regret (with $Omega(D)$ exp terms) on the $D$-chain environment, and that a `polynomial' UCT variant has $exp_2(exp_2(D - O(log D)))$ regret on the same environment -- the original proofs contain an oversight for rewards bounded in $[0, 1]$, which we fix in the present draft. We also adapt the proofs to AlphaGo's MCTS and its descendants (e.g., AlphaZero, Leela Zero) to also show $exp_2(exp_2(D - O(log D)))$ regret. | [
"['Laurent Orseau' 'Remi Munos']"
]
|
null | null | 2405.04437 | null | null | http://arxiv.org/pdf/2405.04437v2 | 2024-07-12T10:33:31Z | 2024-05-07T16:00:32Z | vAttention: Dynamic Memory Management for Serving LLMs without
PagedAttention | Efficient management of GPU memory is essential for high throughput LLM inference. Prior systems used to reserve KV-cache memory ahead-of-time that resulted in wasted capacity due to internal fragmentation. Inspired by demand paging, vLLM proposed PagedAttention to enable dynamic memory allocation for KV-cache. This approach eliminates fragmentation and improves serving throughout. However, to be able to allocate physical memory dynamically, PagedAttention changes the layout of KV-cache from contiguous virtual memory to non-contiguous virtual memory. As a consequence, one needs to rewrite the attention kernels to support paging, and implement a memory manager in the serving framework. This results in both performance and programming overheads, as well as portability challenges in adopting state-of-the-art attention kernels. In this paper, we propose vAttention, a new approach for dynamic KV-cache memory management. In contrast to PagedAttention, vAttention stores KV-cache in contiguous virtual memory and leverages OS support for on-demand allocation of physical memory. vAttention thus enables one to use state-of-the art attention kernels out-of-the-box by adding support for dynamic allocation of physical memory without having to re-write their code. We implement vAttention in the vLLM serving stack to show that it also helps improve decode throughput by up to 1.99x over vLLM, and the end-to-end serving throughput by up to 1.22x and 1.29x, compared to using the state-of-the-art PagedAttention based kernels of FlashAttention and FlashInfer. | [
"['Ramya Prabhu' 'Ajay Nayak' 'Jayashree Mohan' 'Ramachandran Ramjee'\n 'Ashish Panwar']"
]
|
null | null | 2405.04459 | null | null | http://arxiv.org/pdf/2405.04459v1 | 2024-05-07T16:24:03Z | 2024-05-07T16:24:03Z | A Significantly Better Class of Activation Functions Than ReLU Like
Activation Functions | This paper introduces a significantly better class of activation functions than the almost universally used ReLU like and Sigmoidal class of activation functions. Two new activation functions referred to as the Cone and Parabolic-Cone that differ drastically from popular activation functions and significantly outperform these on the CIFAR-10 and Imagenette benchmmarks are proposed. The cone activation functions are positive only on a finite interval and are strictly negative except at the end-points of the interval, where they become zero. Thus the set of inputs that produce a positive output for a neuron with cone activation functions is a hyperstrip and not a half-space as is the usual case. Since a hyper strip is the region between two parallel hyper-planes, it allows neurons to more finely divide the input feature space into positive and negative classes than with infinitely wide half-spaces. In particular the XOR function can be learn by a single neuron with cone-like activation functions. Both the cone and parabolic-cone activation functions are shown to achieve higher accuracies with significantly fewer neurons on benchmarks. The results presented in this paper indicate that many nonlinear real-world datasets may be separated with fewer hyperstrips than half-spaces. The Cone and Parabolic-Cone activation functions have larger derivatives than ReLU and are shown to significantly speedup training. | [
"['Mathew Mithra Noel' 'Yug Oswal']"
]
|
null | null | 2405.04484 | null | null | http://arxiv.org/pdf/2405.04484v1 | 2024-05-07T16:53:29Z | 2024-05-07T16:53:29Z | OptPDE: Discovering Novel Integrable Systems via AI-Human Collaboration | Integrable partial differential equation (PDE) systems are of great interest in natural science, but are exceedingly rare and difficult to discover. To solve this, we introduce OptPDE, a first-of-its-kind machine learning approach that Optimizes PDEs' coefficients to maximize their number of conserved quantities, $n_{rm CQ}$, and thus discover new integrable systems. We discover four families of integrable PDEs, one of which was previously known, and three of which have at least one conserved quantity but are new to the literature to the best of our knowledge. We investigate more deeply the properties of one of these novel PDE families, $u_t = (u_x+a^2u_{xxx})^3$. Our paper offers a promising schema of AI-human collaboration for integrable system discovery: machine learning generates interpretable hypotheses for possible integrable systems, which human scientists can verify and analyze, to truly close the discovery loop. | [
"['Subhash Kantamneni' 'Ziming Liu' 'Max Tegmark']"
]
|
null | null | 2405.04485 | null | null | http://arxiv.org/pdf/2405.04485v1 | 2024-05-07T16:53:42Z | 2024-05-07T16:53:42Z | Adapting WavLM for Speech Emotion Recognition | Recently, the usage of speech self-supervised models (SSL) for downstream tasks has been drawing a lot of attention. While large pre-trained models commonly outperform smaller models trained from scratch, questions regarding the optimal fine-tuning strategies remain prevalent. In this paper, we explore the fine-tuning strategies of the WavLM Large model for the speech emotion recognition task on the MSP Podcast Corpus. More specifically, we perform a series of experiments focusing on using gender and semantic information from utterances. We then sum up our findings and describe the final model we used for submission to Speech Emotion Recognition Challenge 2024. | [
"['Daria Diatlova' 'Anton Udalov' 'Vitalii Shutov' 'Egor Spirin']"
]
|
null | null | 2405.04491 | null | null | http://arxiv.org/pdf/2405.04491v1 | 2024-05-07T17:02:02Z | 2024-05-07T17:02:02Z | TorchDriveEnv: A Reinforcement Learning Benchmark for Autonomous Driving
with Reactive, Realistic, and Diverse Non-Playable Characters | The training, testing, and deployment, of autonomous vehicles requires realistic and efficient simulators. Moreover, because of the high variability between different problems presented in different autonomous systems, these simulators need to be easy to use, and easy to modify. To address these problems we introduce TorchDriveSim and its benchmark extension TorchDriveEnv. TorchDriveEnv is a lightweight reinforcement learning benchmark programmed entirely in Python, which can be modified to test a number of different factors in learned vehicle behavior, including the effect of varying kinematic models, agent types, and traffic control patterns. Most importantly unlike many replay based simulation approaches, TorchDriveEnv is fully integrated with a state of the art behavioral simulation API. This allows users to train and evaluate driving models alongside data driven Non-Playable Characters (NPC) whose initializations and driving behavior are reactive, realistic, and diverse. We illustrate the efficiency and simplicity of TorchDriveEnv by evaluating common reinforcement learning baselines in both training and validation environments. Our experiments show that TorchDriveEnv is easy to use, but difficult to solve. | [
"['Jonathan Wilder Lavington' 'Ke Zhang' 'Vasileios Lioutas'\n 'Matthew Niedoba' 'Yunpeng Liu' 'Dylan Green' 'Saeid Naderiparizi'\n 'Xiaoxuan Liang' 'Setareh Dabiri' 'Adam Ścibior' 'Berend Zwartsenberg'\n 'Frank Wood']"
]
|
null | null | 2405.04494 | null | null | http://arxiv.org/pdf/2405.04494v1 | 2024-05-07T17:04:21Z | 2024-05-07T17:04:21Z | Representation Learning of Daily Movement Data Using Text Encoders | Time-series representation learning is a key area of research for remote healthcare monitoring applications. In this work, we focus on a dataset of recordings of in-home activity from people living with Dementia. We design a representation learning method based on converting activity to text strings that can be encoded using a language model fine-tuned to transform data from the same participants within a $30$-day window to similar embeddings in the vector space. This allows for clustering and vector searching over participants and days, and the identification of activity deviations to aid with personalised delivery of care. | [
"['Alexander Capstick' 'Tianyu Cui' 'Yu Chen' 'Payam Barnaghi']"
]
|
null | null | 2405.04495 | null | null | http://arxiv.org/pdf/2405.04495v1 | 2024-05-07T17:05:27Z | 2024-05-07T17:05:27Z | Toward In-Context Teaching: Adapting Examples to Students'
Misconceptions | When a teacher provides examples for a student to study, these examples must be informative, enabling a student to progress from their current state toward a target concept or skill. Good teachers must therefore simultaneously infer what students already know and adapt their teaching to students' changing state of knowledge. There is increasing interest in using computational models, particularly large language models, as pedagogical tools. As students, language models in particular have shown a remarkable ability to adapt to new tasks given small numbers of examples. But how effectively can these models adapt as teachers to students of different types? To study this question, we introduce a suite of models and evaluation methods we call AdapT. AdapT has two components: (1) a collection of simulated Bayesian student models that can be used for evaluation of automated teaching methods; (2) a platform for evaluation with human students, to characterize the real-world effectiveness of these methods. We additionally introduce (3) AToM, a new probabilistic model for adaptive teaching that jointly infers students' past beliefs and optimizes for the correctness of future beliefs. In evaluations of simulated students across three learning domains (fraction arithmetic, English morphology, function learning), AToM systematically outperforms LLM-based and standard Bayesian teaching models. In human experiments, both AToM and LLMs outperform non-adaptive random example selection. Our results highlight both the difficulty of the adaptive teaching task and the potential of learned adaptive models for solving it. | [
"['Alexis Ross' 'Jacob Andreas']"
]
|
null | null | 2405.04507 | null | null | http://arxiv.org/pdf/2405.04507v1 | 2024-05-07T17:38:39Z | 2024-05-07T17:38:39Z | New allometric models for the USA create a step-change in forest carbon
estimation, modeling, and mapping | The United States national forest inventory (NFI) serves as the foundation for forest aboveground biomass (AGB) and carbon accounting across the nation. These data enable design-based estimates of forest carbon stocks and stock-changes at state and regional levels, but also serve as inputs to model-based approaches for characterizing forest carbon stocks and stock-changes at finer resolutions. Although NFI tree and plot-level data are often treated as truth in these models, they are in fact estimates based on regional species-group models known collectively as the Component Ratio Method (CRM). In late 2023 the Forest Inventory and Analysis (FIA) program introduced a new National Scale Volume and Biomass Estimators (NSVB) system to replace CRM nationwide and offer more precise and accurate representations of forest AGB and carbon. Given the prevalence of model-based AGB studies relying on FIA, there is concern about the transferability of methods from CRM to NSVB models, as well as the comparability of existing CRM AGB products (e.g. maps) to new and forthcoming NSVB AGB products. To begin addressing these concerns we compared previously published CRM AGB maps to new maps produced using identical methods with NSVB AGB reference data. Our results suggest that models relying on passive satellite imagery (e.g. Landsat) provide acceptable estimates of point-in-time NSVB AGB and carbon stocks, but fail to accurately quantify growth in mature closed-canopy forests. We highlight that existing estimates, models, and maps based on FIA reference data are no longer compatible with NSVB, and recommend new methods as well as updated models and maps for accommodating this step-change. Our collective ability to adopt NSVB in our modeling and mapping workflows will help us provide the most accurate spatial forest carbon data possible in order to better inform local management and decision making. | [
"['Lucas K. Johnson' 'Michael J. Mahoney' 'Grant Domke' 'Colin M. Beier']"
]
|
null | null | 2405.04513 | null | null | http://arxiv.org/pdf/2405.04513v1 | 2024-05-07T17:44:54Z | 2024-05-07T17:44:54Z | Switchable Decision: Dynamic Neural Generation Networks | Auto-regressive generation models achieve competitive performance across many different NLP tasks such as summarization, question answering, and classifications. However, they are also known for being slow in inference, which makes them challenging to deploy in real-time applications. We propose a switchable decision to accelerate inference by dynamically assigning computation resources for each data instance. Automatically making decisions on where to skip and how to balance quality and computation cost with constrained optimization, our dynamic neural generation networks enforce the efficient inference path and determine the optimized trade-off. Experiments across question answering, summarization, and classification benchmarks show that our method benefits from less computation cost during inference while keeping the same accuracy. Extensive experiments and ablation studies demonstrate that our method can be general, effective, and beneficial for many NLP tasks. | [
"['Shujian Zhang' 'Korawat Tanwisuth' 'Chengyue Gong' 'Pengcheng He'\n 'Mingyuan Zhou']"
]
|
null | null | 2405.04517 | null | null | http://arxiv.org/pdf/2405.04517v1 | 2024-05-07T17:50:21Z | 2024-05-07T17:50:21Z | xLSTM: Extended Long Short-Term Memory | In the 1990s, the constant error carousel and gating were introduced as the central ideas of the Long Short-Term Memory (LSTM). Since then, LSTMs have stood the test of time and contributed to numerous deep learning success stories, in particular they constituted the first Large Language Models (LLMs). However, the advent of the Transformer technology with parallelizable self-attention at its core marked the dawn of a new era, outpacing LSTMs at scale. We now raise a simple question: How far do we get in language modeling when scaling LSTMs to billions of parameters, leveraging the latest techniques from modern LLMs, but mitigating known limitations of LSTMs? Firstly, we introduce exponential gating with appropriate normalization and stabilization techniques. Secondly, we modify the LSTM memory structure, obtaining: (i) sLSTM with a scalar memory, a scalar update, and new memory mixing, (ii) mLSTM that is fully parallelizable with a matrix memory and a covariance update rule. Integrating these LSTM extensions into residual block backbones yields xLSTM blocks that are then residually stacked into xLSTM architectures. Exponential gating and modified memory structures boost xLSTM capabilities to perform favorably when compared to state-of-the-art Transformers and State Space Models, both in performance and scaling. | [
"['Maximilian Beck' 'Korbinian Pöppel' 'Markus Spanring' 'Andreas Auer'\n 'Oleksandra Prudnikova' 'Michael Kopp' 'Günter Klambauer'\n 'Johannes Brandstetter' 'Sepp Hochreiter']"
]
|
null | null | 2405.04520 | null | null | http://arxiv.org/pdf/2405.04520v1 | 2024-05-07T17:52:51Z | 2024-05-07T17:52:51Z | NaturalCodeBench: Examining Coding Performance Mismatch on HumanEval and
Natural User Prompts | Large language models (LLMs) have manifested strong ability to generate codes for productive activities. However, current benchmarks for code synthesis, such as HumanEval, MBPP, and DS-1000, are predominantly oriented towards introductory tasks on algorithm and data science, insufficiently satisfying challenging requirements prevalent in real-world coding. To fill this gap, we propose NaturalCodeBench (NCB), a challenging code benchmark designed to mirror the complexity and variety of scenarios in real coding tasks. NCB comprises 402 high-quality problems in Python and Java, meticulously selected from natural user queries from online coding services, covering 6 different domains. Noting the extraordinary difficulty in creating testing cases for real-world queries, we also introduce a semi-automated pipeline to enhance the efficiency of test case construction. Comparing with manual solutions, it achieves an efficiency increase of more than 4 times. Our systematic experiments on 39 LLMs find that performance gaps on NCB between models with close HumanEval scores could still be significant, indicating a lack of focus on practical code synthesis scenarios or over-specified optimization on HumanEval. On the other hand, even the best-performing GPT-4 is still far from satisfying on NCB. The evaluation toolkit and development set are available at https://github.com/THUDM/NaturalCodeBench. | [
"['Shudan Zhang' 'Hanlin Zhao' 'Xiao Liu' 'Qinkai Zheng' 'Zehan Qi'\n 'Xiaotao Gu' 'Xiaohan Zhang' 'Yuxiao Dong' 'Jie Tang']"
]
|
null | null | 2405.04532 | null | null | http://arxiv.org/pdf/2405.04532v2 | 2024-05-10T15:58:26Z | 2024-05-07T17:59:30Z | QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM
Serving | Quantization can accelerate large language model (LLM) inference. Going beyond INT8 quantization, the research community is actively exploring even lower precision, such as INT4. Nonetheless, state-of-the-art INT4 quantization techniques only accelerate low-batch, edge LLM inference, failing to deliver performance gains in large-batch, cloud-based LLM serving. We uncover a critical issue: existing INT4 quantization methods suffer from significant runtime overhead (20-90%) when dequantizing either weights or partial sums on GPUs. To address this challenge, we introduce QoQ, a W4A8KV4 quantization algorithm with 4-bit weight, 8-bit activation, and 4-bit KV cache. QoQ stands for quattuor-octo-quattuor, which represents 4-8-4 in Latin. QoQ is implemented by the QServe inference library that achieves measured speedup. The key insight driving QServe is that the efficiency of LLM serving on GPUs is critically influenced by operations on low-throughput CUDA cores. Building upon this insight, in QoQ algorithm, we introduce progressive quantization that can allow low dequantization overhead in W4A8 GEMM. Additionally, we develop SmoothAttention to effectively mitigate the accuracy degradation incurred by 4-bit KV quantization. In the QServe system, we perform compute-aware weight reordering and take advantage of register-level parallelism to reduce dequantization latency. We also make fused attention memory-bound, harnessing the performance gain brought by KV4 quantization. As a result, QServe improves the maximum achievable serving throughput of Llama-3-8B by 1.2x on A100, 1.4x on L40S; and Qwen1.5-72B by 2.4x on A100, 3.5x on L40S, compared to TensorRT-LLM. Remarkably, QServe on L40S GPU can achieve even higher throughput than TensorRT-LLM on A100. Thus, QServe effectively reduces the dollar cost of LLM serving by 3x. Code is available at https://github.com/mit-han-lab/qserve. | [
"['Yujun Lin' 'Haotian Tang' 'Shang Yang' 'Zhekai Zhang' 'Guangxuan Xiao'\n 'Chuang Gan' 'Song Han']"
]
|
null | null | 2405.04533 | null | null | http://arxiv.org/pdf/2405.04533v1 | 2024-05-07T17:59:31Z | 2024-05-07T17:59:31Z | ChatHuman: Language-driven 3D Human Understanding with
Retrieval-Augmented Tool Reasoning | Numerous methods have been proposed to detect, estimate, and analyze properties of people in images, including the estimation of 3D pose, shape, contact, human-object interaction, emotion, and more. Each of these methods works in isolation instead of synergistically. Here we address this problem and build a language-driven human understanding system -- ChatHuman, which combines and integrates the skills of many different methods. To do so, we finetune a Large Language Model (LLM) to select and use a wide variety of existing tools in response to user inputs. In doing so, ChatHuman is able to combine information from multiple tools to solve problems more accurately than the individual tools themselves and to leverage tool output to improve its ability to reason about humans. The novel features of ChatHuman include leveraging academic publications to guide the application of 3D human-related tools, employing a retrieval-augmented generation model to generate in-context-learning examples for handling new tools, and discriminating and integrating tool results to enhance 3D human understanding. Our experiments show that ChatHuman outperforms existing models in both tool selection accuracy and performance across multiple 3D human-related tasks. ChatHuman is a step towards consolidating diverse methods for human analysis into a single, powerful, system for 3D human reasoning. | [
"['Jing Lin' 'Yao Feng' 'Weiyang Liu' 'Michael J. Black']"
]
|
null | null | 2405.04536 | null | null | http://arxiv.org/pdf/2405.04536v1 | 2024-03-15T10:12:45Z | 2024-03-15T10:12:45Z | When Training-Free NAS Meets Vision Transformer: A Neural Tangent Kernel
Perspective | This paper investigates the Neural Tangent Kernel (NTK) to search vision transformers without training. In contrast with the previous observation that NTK-based metrics can effectively predict CNNs performance at initialization, we empirically show their inefficacy in the ViT search space. We hypothesize that the fundamental feature learning preference within ViT contributes to the ineffectiveness of applying NTK to NAS for ViT. We both theoretically and empirically validate that NTK essentially estimates the ability of neural networks that learn low-frequency signals, completely ignoring the impact of high-frequency signals in feature learning. To address this limitation, we propose a new method called ViNTK that generalizes the standard NTK to the high-frequency domain by integrating the Fourier features from inputs. Experiments with multiple ViT search spaces on image classification and semantic segmentation tasks show that our method can significantly speed up search costs over prior state-of-the-art NAS for ViT while maintaining similar performance on searched architectures. | [
"['Qiqi Zhou' 'Yichen Zhu']"
]
|
null | null | 2405.04538 | null | null | http://arxiv.org/pdf/2405.04538v1 | 2024-03-15T14:34:29Z | 2024-03-15T14:34:29Z | DiffFinger: Advancing Synthetic Fingerprint Generation through Denoising
Diffusion Probabilistic Models | This study explores the generation of synthesized fingerprint images using Denoising Diffusion Probabilistic Models (DDPMs). The significant obstacles in collecting real biometric data, such as privacy concerns and the demand for diverse datasets, underscore the imperative for synthetic biometric alternatives that are both realistic and varied. Despite the strides made with Generative Adversarial Networks (GANs) in producing realistic fingerprint images, their limitations prompt us to propose DDPMs as a promising alternative. DDPMs are capable of generating images with increasing clarity and realism while maintaining diversity. Our results reveal that DiffFinger not only competes with authentic training set data in quality but also provides a richer set of biometric data, reflecting true-to-life variability. These findings mark a promising stride in biometric synthesis, showcasing the potential of DDPMs to advance the landscape of fingerprint identification and authentication systems. | [
"['Freddie Grabovski' 'Lior Yasur' 'Yaniv Hacmon' 'Lior Nisimov'\n 'Stav Nimrod']"
]
|
null | null | 2405.04539 | null | null | http://arxiv.org/pdf/2405.04539v1 | 2024-04-07T17:41:02Z | 2024-04-07T17:41:02Z | Some variation of COBRA in sequential learning setup | This research paper introduces innovative approaches for multivariate time series forecasting based on different variations of the combined regression strategy. We use specific data preprocessing techniques which makes a radical change in the behaviour of prediction. We compare the performance of the model based on two types of hyper-parameter tuning Bayesian optimisation (BO) and Usual Grid search. Our proposed methodologies outperform all state-of-the-art comparative models. We illustrate the methodologies through eight time series datasets from three categories: cryptocurrency, stock index, and short-term load forecasting. | [
"['Aryan Bhambu' 'Arabin Kumar Dey']"
]
|
null | null | 2405.04545 | null | null | http://arxiv.org/pdf/2405.04545v1 | 2024-05-03T21:18:43Z | 2024-05-03T21:18:43Z | Learning label-label correlations in Extreme Multi-label Classification
via Label Features | Extreme Multi-label Text Classification (XMC) involves learning a classifier that can assign an input with a subset of most relevant labels from millions of label choices. Recent works in this domain have increasingly focused on a symmetric problem setting where both input instances and label features are short-text in nature. Short-text XMC with label features has found numerous applications in areas such as query-to-ad-phrase matching in search ads, title-based product recommendation, prediction of related searches. In this paper, we propose Gandalf, a novel approach which makes use of a label co-occurrence graph to leverage label features as additional data points to supplement the training distribution. By exploiting the characteristics of the short-text XMC problem, it leverages the label features to construct valid training instances, and uses the label graph for generating the corresponding soft-label targets, hence effectively capturing the label-label correlations. Surprisingly, models trained on these new training instances, although being less than half of the original dataset, can outperform models trained on the original dataset, particularly on the PSP@k metric for tail labels. With this insight, we aim to train existing XMC algorithms on both, the original and new training instances, leading to an average 5% relative improvements for 6 state-of-the-art algorithms across 4 benchmark datasets consisting of up to 1.3M labels. Gandalf can be applied in a plug-and-play manner to various methods and thus forwards the state-of-the-art in the domain, without incurring any additional computational overheads. | [
"['Siddhant Kharbanda' 'Devaansh Gupta' 'Erik Schultheis'\n 'Atmadeep Banerjee' 'Cho-Jui Hsieh' 'Rohit Babbar']"
]
|
null | null | 2405.04551 | null | null | http://arxiv.org/pdf/2405.04551v2 | 2024-06-04T05:17:56Z | 2024-05-06T03:19:24Z | Differentially Private Federated Learning without Noise Addition: When
is it Possible? | Federated Learning (FL) with Secure Aggregation (SA) has gained significant attention as a privacy preserving framework for training machine learning models while preventing the server from learning information about users' data from their individual encrypted model updates. Recent research has extended privacy guarantees of FL with SA by bounding the information leakage through the aggregate model over multiple training rounds thanks to leveraging the "noise" from other users' updates. However, the privacy metric used in that work (mutual information) measures the on-average privacy leakage, without providing any privacy guarantees for worse-case scenarios. To address this, in this work we study the conditions under which FL with SA can provide worst-case differential privacy guarantees. Specifically, we formally identify the necessary condition that SA can provide DP without addition noise. We then prove that when the randomness inside the aggregated model update is Gaussian with non-singular covariance matrix, SA can provide differential privacy guarantees with the level of privacy $epsilon$ bounded by the reciprocal of the minimum eigenvalue of the covariance matrix. However, we further demonstrate that in practice, these conditions are almost unlikely to hold and hence additional noise added in model updates is still required in order for SA in FL to achieve DP. Lastly, we discuss the potential solution of leveraging inherent randomness inside aggregated model update to reduce the amount of addition noise required for DP guarantee. | [
"['Jiang Zhang' 'Konstantinos Psounis']"
]
|
null | null | 2405.04554 | null | null | http://arxiv.org/pdf/2405.04554v1 | 2024-05-06T14:06:12Z | 2024-05-06T14:06:12Z | Differentially Private Synthetic Data with Private Density Estimation | The need to analyze sensitive data, such as medical records or financial data, has created a critical research challenge in recent years. In this paper, we adopt the framework of differential privacy, and explore mechanisms for generating an entire dataset which accurately captures characteristics of the original data. We build upon the work of Boedihardjo et al, which laid the foundations for a new optimization-based algorithm for generating private synthetic data. Importantly, we adapt their algorithm by replacing a uniform sampling step with a private distribution estimator; this allows us to obtain better computational guarantees for discrete distributions, and develop a novel algorithm suitable for continuous distributions. We also explore applications of our work to several statistical tasks. | [
"['Nikolija Bojkovic' 'Po-Ling Loh']"
]
|
null | null | 2405.04561 | null | null | http://arxiv.org/abs/2405.04561v1 | 2024-05-07T14:54:32Z | 2024-05-07T14:54:32Z | Inferring Discussion Topics about Exploitation of Vulnerabilities from
Underground Hacking Forums | The increasing sophistication of cyber threats necessitates proactive measures to identify vulnerabilities and potential exploits. Underground hacking forums serve as breeding grounds for the exchange of hacking techniques and discussions related to exploitation. In this research, we propose an innovative approach using topic modeling to analyze and uncover key themes in vulnerabilities discussed within these forums. The objective of our study is to develop a machine learning-based model that can automatically detect and classify vulnerability-related discussions in underground hacking forums. By monitoring and analyzing the content of these forums, we aim to identify emerging vulnerabilities, exploit techniques, and potential threat actors. To achieve this, we collect a large-scale dataset consisting of posts and threads from multiple underground forums. We preprocess and clean the data to ensure accuracy and reliability. Leveraging topic modeling techniques, specifically Latent Dirichlet Allocation (LDA), we uncover latent topics and their associated keywords within the dataset. This enables us to identify recurring themes and prevalent discussions related to vulnerabilities, exploits, and potential targets. | [
"['Felipe Moreno-Vera']"
]
|
null | null | 2405.04566 | null | null | http://arxiv.org/pdf/2405.04566v1 | 2024-05-07T17:25:56Z | 2024-05-07T17:25:56Z | Fast Decentralized Gradient Tracking for Federated Minimax Optimization
with Local Updates | Federated learning (FL) for minimax optimization has emerged as a powerful paradigm for training models across distributed nodes/clients while preserving data privacy and model robustness on data heterogeneity. In this work, we delve into the decentralized implementation of federated minimax optimization by proposing texttt{K-GT-Minimax}, a novel decentralized minimax optimization algorithm that combines local updates and gradient tracking techniques. Our analysis showcases the algorithm's communication efficiency and convergence rate for nonconvex-strongly-concave (NC-SC) minimax optimization, demonstrating a superior convergence rate compared to existing methods. texttt{K-GT-Minimax}'s ability to handle data heterogeneity and ensure robustness underscores its significance in advancing federated learning research and applications. | [
"['Chris Junchi Li']"
]
|
null | null | 2405.04579 | null | null | http://arxiv.org/pdf/2405.04579v2 | 2024-06-09T06:44:06Z | 2024-04-30T09:05:17Z | A critical appraisal of water table depth estimation: Challenges and
opportunities within machine learning | Fine-resolution spatial patterns of water table depth (WTD) play a crucial role in shaping ecological resilience, hydrological connectivity, and anthropocentric objectives. Generally, a large-scale (e.g., continental or global) spatial map of static WTD can be simulated using either physically-based (PB) or machine learning-based (ML) models. We construct three fine-resolution (500 m) ML simulations of WTD, using the XGBoost algorithm and more than 20 million real and proxy observations of WTD, across the United States and Canada. The three ML models were constrained using known physical relations between WTD's drivers and WTD and were trained by sequentially adding real and proxy observations of WTD. We interpret the black box of our physically constrained ML models and compare it against available literature in groundwater hydrology. Through an extensive (pixel-by-pixel) evaluation, we demonstrate that our models can more accurately predict unseen real and proxy observations of WTD across most of North America's ecoregions compared to three available PB simulations of WTD. However, we still argue that large-scale WTD estimation is far from being a solved problem. We reason that due to biased observational data mainly collected from low-elevation floodplains, the misspecification of equations within physically-based models, and the over-flexibility of machine learning models, verifiably accurate simulations of WTD do not yet exist. Ultimately, we thoroughly discuss future directions that may help hydrogeologists decide how to proceed with WTD estimations, with a particular focus on the application of machine learning and the use of proxy satellite data. | [
"['Joseph Janssen' 'Ardalan Tootchi' 'Ali A. Ameli']"
]
|
null | null | 2405.04585 | null | null | http://arxiv.org/pdf/2405.04585v1 | 2024-04-29T10:30:59Z | 2024-04-29T10:30:59Z | PoPE: Legendre Orthogonal Polynomials Based Position Encoding for Large
Language Models | There are several improvements proposed over the baseline Absolute Positional Encoding (APE) method used in original transformer. In this study, we aim to investigate the implications of inadequately representing positional encoding in higher dimensions on crucial aspects of the attention mechanism, the model's capacity to learn relative positional information, and the convergence of models, all stemming from the choice of sinusoidal basis functions. Through a combination of theoretical insights and empirical analyses, we elucidate how these challenges extend beyond APEs and may adversely affect the performance of Relative Positional Encoding (RPE) methods, such as Rotatory Positional Encoding (RoPE). Subsequently, we introduce an innovative solution termed Orthogonal Polynomial Based Positional Encoding (PoPE) to address some of the limitations associated with existing methods. The PoPE method encodes positional information by leveraging Orthogonal Legendre polynomials. Legendre polynomials as basis functions offers several desirable properties for positional encoding, including improved correlation structure, non-periodicity, orthogonality, and distinct functional forms among polynomials of varying orders. Our experimental findings demonstrate that transformer models incorporating PoPE outperform baseline transformer models on the $Multi30k$ English-to-German translation task, thus establishing a new performance benchmark. Furthermore, PoPE-based transformers exhibit significantly accelerated convergence rates. Additionally, we will present novel theoretical perspectives on position encoding based on the superior performance of PoPE. | [
"['Arpit Aggarwal']"
]
|
null | null | 2405.04592 | null | null | http://arxiv.org/pdf/2405.04592v1 | 2024-05-07T18:10:54Z | 2024-05-07T18:10:54Z | Integrating knowledge-guided symbolic regression and model-based design
of experiments to automate process flow diagram development | New products must be formulated rapidly to succeed in the global formulated product market; however, key product indicators (KPIs) can be complex, poorly understood functions of the chemical composition and processing history. Consequently, scale-up must currently undergo expensive trial-and-error campaigns. To accelerate process flow diagram (PFD) optimisation and knowledge discovery, this work proposed a novel digital framework to automatically quantify process mechanisms by integrating symbolic regression (SR) within model-based design of experiments (MBDoE). Each iteration, SR proposed a Pareto front of interpretable mechanistic expressions, and then MBDoE designed a new experiment to discriminate between them while balancing PFD optimisation. To investigate the framework's performance, a new process model capable of simulating general formulated product synthesis was constructed to generate in-silico data for different case studies. The framework could effectively discover ground-truth process mechanisms within a few iterations, indicating its great potential for use within the general chemical industry for digital manufacturing and product innovation. | [
"['Alexander W. Rogers' 'Amanda Lane' 'Cesar Mendoza' 'Simon Watson'\n 'Adam Kowalski' 'Philip Martin' 'Dongda Zhang']"
]
|
null | null | 2405.04605 | null | null | http://arxiv.org/pdf/2405.04605v2 | 2024-06-12T22:18:41Z | 2024-05-07T18:36:40Z | AI in Lung Health: Benchmarking Detection and Diagnostic Models Across
Multiple CT Scan Datasets | Lung cancer's high mortality rate can be mitigated by early detection, increasingly reliant on AI for diagnostic imaging. However, AI model performance depends on training and validation datasets. This study develops and validates AI models for both nodule detection and cancer classification tasks. For detection, two models (DLCSD-mD and LUNA16-mD) were developed using the Duke Lung Cancer Screening Dataset (DLCSD), with over 2,000 CT scans from 1,613 patients and more than 3,000 annotations. These models were evaluated on internal (DLCSD) and external datasets, including LUNA16 (601 patients, 1186 nodules) and NLST (969 patients, 1192 nodules), using FROC analysis and AUC metrics. For classification, five models were developed and tested: a randomly initialized 3D ResNet50, Genesis, MedNet3D, an enhanced ResNet50 using Strategic Warm-Start++ (SWS++), and a linear classifier analyzing features from the Foundation Model for Cancer Biomarkers (FMCB). These models were trained to distinguish between benign and malignant nodules and evaluated using AUC analysis on internal (DLCSD) and external datasets, including LUNA16 (433 patients, 677 nodules) and NLST. The DLCSD-mD model achieved an AUC of 0.93 (95% CI: 0.91-0.94) on the internal DLCSD dataset. External validation results were 0.97 (95% CI: 0.96-0.98) on LUNA16 and 0.75 (95% CI: 0.73-0.76) on NLST. For classification, the ResNet50-SWS++ model recorded AUCs of 0.71 (95% CI: 0.61-0.81) on DLCSD, 0.90 (95% CI: 0.87-0.93) on LUNA16, and 0.81 (95% CI: 0.79-0.82) on NLST. Other models showed varying performance across datasets, underscoring the importance of diverse model approaches. This benchmarking establishes DLCSD as a reliable resource for lung cancer AI research. | [
"['Fakrul Islam Tushar' 'Avivah Wang' 'Lavsen Dahal' 'Michael R. Harowicz'\n 'Kyle J. Lafata' 'Tina D. Tailor' 'Joseph Y. Lo']"
]
|
null | null | 2405.04614 | null | null | http://arxiv.org/pdf/2405.04614v2 | 2024-06-23T22:05:46Z | 2024-05-07T18:58:32Z | Multi-Margin Loss: Proposal and Application in Recommender Systems | Recommender systems guide users through vast amounts of information by suggesting items based on their predicted preferences. Collaborative filtering-based deep learning techniques have regained popularity due to their simplicity, using only user-item interactions. Typically, these systems consist of three main components: an interaction module, a loss function, and a negative sampling strategy. Initially, researchers focused on enhancing performance by developing complex interaction modules with techniques like multi-layer perceptrons, transformers, or graph neural networks. However, there has been a recent shift toward refining loss functions and negative sampling strategies. This shift has increased interest in contrastive learning, which pulls similar pairs closer while pushing dissimilar ones apart. Contrastive learning involves key practices such as heavy data augmentation, large batch sizes, and hard-negative sampling, but these also bring challenges like high memory demands and under-utilization of some negative samples. The proposed Multi-Margin Loss (MML) addresses these challenges by introducing multiple margins and varying weights for negative samples. MML efficiently utilizes not only the hardest negatives but also other non-trivial negatives, offering a simpler yet effective loss function that outperforms more complex methods, especially when resources are limited. Experiments on two well-known datasets showed MML achieved up to a 20% performance improvement compared to a baseline contrastive loss function with fewer negative samples. | [
"['Makbule Gulcin Ozsoy']"
]
|
null | null | 2405.04620 | null | null | http://arxiv.org/pdf/2405.04620v2 | 2024-05-10T02:18:27Z | 2024-05-07T19:05:26Z | Folded context condensation in Path Integral formalism for infinite
context transformers | This short note is written for rapid communication of long context training and to share the idea of how to train it with low memory usage. In the note, we generalize the attention algorithm and neural network of Generative Pre-Trained Transformers and reinterpret it in Path integral formalism. First, the role of the transformer is understood as the time evolution of the token state and second, it is suggested that the all key-token states in the same time as the query-token can attend to the attention with the query token states. As a result of the repetitive time evolution, it is discussed that the token states in the past sequence meats the token states in the present sequence so that the attention between separated sequences becomes possible for maintaining infinite contextual information just by using low memory for limited size of sequence. For the experiment, the $12$ input token window size was taken and one GPU with $24$GB memory was used for the pre-training. It was confirmed that more than $150$ length context is preserved. The sampling result of the training, the code and the other details will be included in the revised version of this note later. | [
"['Won-Gi Paeng' 'Daesuk Kwon']"
]
|
null | null | 2405.04634 | null | null | http://arxiv.org/pdf/2405.04634v2 | 2024-05-13T10:07:10Z | 2024-05-07T19:37:22Z | FRACTAL: An Ultra-Large-Scale Aerial Lidar Dataset for 3D Semantic
Segmentation of Diverse Landscapes | Mapping agencies are increasingly adopting Aerial Lidar Scanning (ALS) as a new tool to monitor territory and support public policies. Processing ALS data at scale requires efficient point classification methods that perform well over highly diverse territories. To evaluate them, researchers need large annotated Lidar datasets, however, current Lidar benchmark datasets have restricted scope and often cover a single urban area. To bridge this data gap, we present the FRench ALS Clouds from TArgeted Landscapes (FRACTAL) dataset: an ultra-large-scale aerial Lidar dataset made of 100,000 dense point clouds with high-quality labels for 7 semantic classes and spanning 250 km$^2$. FRACTAL is built upon France's nationwide open Lidar data. It achieves spatial and semantic diversity via a sampling scheme that explicitly concentrates rare classes and challenging landscapes from five French regions. It should support the development of 3D deep learning approaches for large-scale land monitoring. We describe the nature of the source data, the sampling workflow, the content of the resulting dataset, and provide an initial evaluation of segmentation performance using a performant 3D neural architecture. | [
"['Charles Gaydon' 'Michel Daab' 'Floryne Roche']"
]
|
null | null | 2405.04636 | null | null | http://arxiv.org/pdf/2405.04636v1 | 2024-05-07T19:38:26Z | 2024-05-07T19:38:26Z | Data-driven Error Estimation: Upper Bounding Multiple Errors with No
Technical Debt | We formulate the problem of constructing multiple simultaneously valid confidence intervals (CIs) as estimating a high probability upper bound on the maximum error for a class/set of estimate-estimand-error tuples, and refer to this as the error estimation problem. For a single such tuple, data-driven confidence intervals can often be used to bound the error in our estimate. However, for a class of estimate-estimand-error tuples, nontrivial high probability upper bounds on the maximum error often require class complexity as input -- limiting the practicality of such methods and often resulting in loose bounds. Rather than deriving theoretical class complexity-based bounds, we propose a completely data-driven approach to estimate an upper bound on the maximum error. The simple and general nature of our solution to this fundamental challenge lends itself to several applications including: multiple CI construction, multiple hypothesis testing, estimating excess risk bounds (a fundamental measure of uncertainty in machine learning) for any training/fine-tuning algorithm, and enabling the development of a contextual bandit pipeline that can leverage any reward model estimation procedure as input (without additional mathematical analysis). | [
"['Sanath Kumar Krishnamurthy' 'Susan Athey' 'Emma Brunskill']"
]
|
null | null | 2405.04657 | null | null | http://arxiv.org/pdf/2405.04657v2 | 2024-06-03T08:50:31Z | 2024-05-07T20:30:14Z | ACEGEN: Reinforcement learning of generative chemical agents for drug
discovery | In recent years, reinforcement learning (RL) has emerged as a valuable tool in drug design, offering the potential to propose and optimize molecules with desired properties. However, striking a balance between capabilities, flexibility, reliability, and efficiency remains challenging due to the complexity of advanced RL algorithms and the significant reliance on specialized code. In this work, we introduce ACEGEN, a comprehensive and streamlined toolkit tailored for generative drug design, built using TorchRL, a modern RL library that offers thoroughly tested reusable components. We validate ACEGEN by benchmarking against other published generative modeling algorithms and show comparable or improved performance. We also show examples of ACEGEN applied in multiple drug discovery case studies. ACEGEN is accessible at url{https://github.com/acellera/acegen-open} and available for use under the MIT license. | [
"['Albert Bou' 'Morgan Thomas' 'Sebastian Dittert' 'Carles Navarro Ramírez'\n 'Maciej Majewski' 'Ye Wang' 'Shivam Patel' 'Gary Tresadern' 'Mazen Ahmad'\n 'Vincent Moens' 'Woody Sherman' 'Simone Sciabola' 'Gianni De Fabritiis']"
]
|
null | null | 2405.04664 | null | null | http://arxiv.org/pdf/2405.04664v1 | 2024-05-07T20:51:49Z | 2024-05-07T20:51:49Z | Proximal Policy Optimization with Adaptive Exploration | Proximal Policy Optimization with Adaptive Exploration (axPPO) is introduced as a novel learning algorithm. This paper investigates the exploration-exploitation tradeoff within the context of reinforcement learning and aims to contribute new insights into reinforcement learning algorithm design. The proposed adaptive exploration framework dynamically adjusts the exploration magnitude during training based on the recent performance of the agent. Our proposed method outperforms standard PPO algorithms in learning efficiency, particularly when significant exploratory behavior is needed at the beginning of the learning process. | [
"['Andrei Lixandru']"
]
|
null | null | 2405.04669 | null | null | http://arxiv.org/pdf/2405.04669v1 | 2024-05-07T21:03:51Z | 2024-05-07T21:03:51Z | Towards a Theoretical Understanding of the 'Reversal Curse' via Training
Dynamics | Auto-regressive large language models (LLMs) show impressive capacities to solve many complex reasoning tasks while struggling with some simple logical reasoning tasks such as inverse search: when trained on ''A is B'', LLM fails to directly conclude ''B is A'' during inference, which is known as the ''reversal curse'' (Berglund et al., 2023). In this paper, we theoretically analyze the reversal curse via the training dynamics of (stochastic) gradient descent for two auto-regressive models: (1) a bilinear model that can be viewed as a simplification of a one-layer transformer; (2) one-layer transformers using the framework of Tian et al. (2023a). Our analysis reveals a core reason why the reversal curse happens: the (effective) weights of both auto-regressive models show asymmetry, i.e., the increase of weights from a token $A$ to token $B$ during training does not necessarily cause the increase of the weights from $B$ to $A$. Moreover, our analysis can be naturally applied to other logical reasoning tasks such as chain-of-thought (COT) (Wei et al., 2022b). We show the necessity of COT, i.e., a model trained on ''$A to B$'' and ''$B to C$'' fails to directly conclude ''$A to C$'' without COT (also empirically observed by Allen-Zhu and Li (2023)), for one-layer transformers via training dynamics, which provides a new perspective different from previous work (Feng et al., 2024) that focuses on expressivity. Finally, we also conduct experiments to validate our theory on multi-layer transformers under different settings. | [
"['Hanlin Zhu' 'Baihe Huang' 'Shaolun Zhang' 'Michael Jordan'\n 'Jiantao Jiao' 'Yuandong Tian' 'Stuart Russell']"
]
|
null | null | 2405.04671 | null | null | http://arxiv.org/pdf/2405.04671v1 | 2024-05-07T21:05:50Z | 2024-05-07T21:05:50Z | Interpretable Tensor Fusion | Conventional machine learning methods are predominantly designed to predict outcomes based on a single data type. However, practical applications may encompass data of diverse types, such as text, images, and audio. We introduce interpretable tensor fusion (InTense), a multimodal learning method for training neural networks to simultaneously learn multimodal data representations and their interpretable fusion. InTense can separately capture both linear combinations and multiplicative interactions of diverse data types, thereby disentangling higher-order interactions from the individual effects of each modality. InTense provides interpretability out of the box by assigning relevance scores to modalities and their associations. The approach is theoretically grounded and yields meaningful relevance scores on multiple synthetic and real-world datasets. Experiments on six real-world datasets show that InTense outperforms existing state-of-the-art multimodal interpretable approaches in terms of accuracy and interpretability. | [
"['Saurabh Varshneya' 'Antoine Ledent' 'Philipp Liznerski'\n 'Andriy Balinskyy' 'Purvanshi Mehta' 'Waleed Mustafa' 'Marius Kloft']"
]
|
null | null | 2405.04682 | null | null | http://arxiv.org/pdf/2405.04682v3 | 2024-05-25T01:13:26Z | 2024-05-07T21:52:39Z | TALC: Time-Aligned Captions for Multi-Scene Text-to-Video Generation | Recent advances in diffusion-based generative modeling have led to the development of text-to-video (T2V) models that can generate high-quality videos conditioned on a text prompt. Most of these T2V models often produce single-scene video clips that depict an entity performing a particular action (e.g., 'a red panda climbing a tree'). However, it is pertinent to generate multi-scene videos since they are ubiquitous in the real-world (e.g., 'a red panda climbing a tree' followed by 'the red panda sleeps on the top of the tree'). To generate multi-scene videos from a pretrained T2V model, we introduce Time-Aligned Captions (TALC) framework. Specifically, we enhance the text-conditioning mechanism in the T2V architecture to recognize the temporal alignment between the video scenes and scene descriptions. As a result, we show that the pretrained T2V model can generate multi-scene videos that adhere to the multi-scene text descriptions and be visually consistent (e.g., w.r.t entity and background). Our TALC-finetuned model outperforms the baseline methods on multi-scene video-text data by 15.5 points on aggregated score, averaging visual consistency and text adherence using human evaluation. The project website is https://talc-mst2v.github.io/. | [
"['Hritik Bansal' 'Yonatan Bitton' 'Michal Yarom' 'Idan Szpektor'\n 'Aditya Grover' 'Kai-Wei Chang']"
]
|
null | null | 2405.04685 | null | null | http://arxiv.org/pdf/2405.04685v1 | 2024-05-07T21:58:45Z | 2024-05-07T21:58:45Z | Bridging the Bosphorus: Advancing Turkish Large Language Models through
Strategies for Low-Resource Language Adaptation and Benchmarking | Large Language Models (LLMs) are becoming crucial across various fields, emphasizing the urgency for high-quality models in underrepresented languages. This study explores the unique challenges faced by low-resource languages, such as data scarcity, model selection, evaluation, and computational limitations, with a special focus on Turkish. We conduct an in-depth analysis to evaluate the impact of training strategies, model choices, and data availability on the performance of LLMs designed for underrepresented languages. Our approach includes two methodologies: (i) adapting existing LLMs originally pretrained in English to understand Turkish, and (ii) developing a model from the ground up using Turkish pretraining data, both supplemented with supervised fine-tuning on a novel Turkish instruction-tuning dataset aimed at enhancing reasoning capabilities. The relative performance of these methods is evaluated through the creation of a new leaderboard for Turkish LLMs, featuring benchmarks that assess different reasoning and knowledge skills. Furthermore, we conducted experiments on data and model scaling, both during pretraining and fine-tuning, simultaneously emphasizing the capacity for knowledge transfer across languages and addressing the challenges of catastrophic forgetting encountered during fine-tuning on a different language. Our goal is to offer a detailed guide for advancing the LLM framework in low-resource linguistic contexts, thereby making natural language processing (NLP) benefits more globally accessible. | [
"['Emre Can Acikgoz' 'Mete Erdogan' 'Deniz Yuret']"
]
|
null | null | 2405.04691 | null | null | http://arxiv.org/pdf/2405.04691v1 | 2024-05-07T22:06:24Z | 2024-05-07T22:06:24Z | Carbon Filter: Real-time Alert Triage Using Large Scale Clustering and
Fast Search | "Alert fatigue" is one of the biggest challenges faced by the Security Operations Center (SOC) today, with analysts spending more than half of their time reviewing false alerts. Endpoint detection products raise alerts by pattern matching on event telemetry against behavioral rules that describe potentially malicious behavior, but can suffer from high false positives that distract from actual attacks. While alert triage techniques based on data provenance may show promise, these techniques can take over a minute to inspect a single alert, while EDR customers may face tens of millions of alerts per day; the current reality is that these approaches aren't nearly scalable enough for production environments. We present Carbon Filter, a statistical learning based system that dramatically reduces the number of alerts analysts need to manually review. Our approach is based on the observation that false alert triggers can be efficiently identified and separated from suspicious behaviors by examining the process initiation context (e.g., the command line) that launched the responsible process. Through the use of fast-search algorithms for training and inference, our approach scales to millions of alerts per day. Through batching queries to the model, we observe a theoretical maximum throughput of 20 million alerts per hour. Based on the analysis of tens of million alerts from customer deployments, our solution resulted in a 6-fold improvement in the Signal-to-Noise ratio without compromising on alert triage performance. | [
"['Jonathan Oliver' 'Raghav Batta' 'Adam Bates' 'Muhammad Adil Inam'\n 'Shelly Mehta' 'Shugao Xia']"
]
|
null | null | 2405.04700 | null | null | http://arxiv.org/pdf/2405.04700v1 | 2024-05-07T22:31:50Z | 2024-05-07T22:31:50Z | Robust Implementation of Retrieval-Augmented Generation on Edge-based
Computing-in-Memory Architectures | Large Language Models (LLMs) deployed on edge devices learn through fine-tuning and updating a certain portion of their parameters. Although such learning methods can be optimized to reduce resource utilization, the overall required resources remain a heavy burden on edge devices. Instead, Retrieval-Augmented Generation (RAG), a resource-efficient LLM learning method, can improve the quality of the LLM-generated content without updating model parameters. However, the RAG-based LLM may involve repetitive searches on the profile data in every user-LLM interaction. This search can lead to significant latency along with the accumulation of user data. Conventional efforts to decrease latency result in restricting the size of saved user data, thus reducing the scalability of RAG as user data continuously grows. It remains an open question: how to free RAG from the constraints of latency and scalability on edge devices? In this paper, we propose a novel framework to accelerate RAG via Computing-in-Memory (CiM) architectures. It accelerates matrix multiplications by performing in-situ computation inside the memory while avoiding the expensive data transfer between the computing unit and memory. Our framework, Robust CiM-backed RAG (RoCR), utilizing a novel contrastive learning-based training method and noise-aware training, can enable RAG to efficiently search profile data with CiM. To the best of our knowledge, this is the first work utilizing CiM to accelerate RAG. | [
"['Ruiyang Qin' 'Zheyu Yan' 'Dewen Zeng' 'Zhenge Jia' 'Dancheng Liu'\n 'Jianbo Liu' 'Zhi Zheng' 'Ningyuan Cao' 'Kai Ni' 'Jinjun Xiong'\n 'Yiyu Shi']"
]
|
null | null | 2405.04710 | null | null | http://arxiv.org/pdf/2405.04710v1 | 2024-05-07T23:08:24Z | 2024-05-07T23:08:24Z | Untangling Lariats: Subgradient Following of Variationally Penalized
Objectives | We describe a novel subgradient following apparatus for calculating the optimum of convex problems with variational penalties. In this setting, we receive a sequence $y_i,ldots,y_n$ and seek a smooth sequence $x_1,ldots,x_n$. The smooth sequence attains the minimum Bregman divergence to an input sequence with additive variational penalties in the general form of $sum_i g_i(x_{i+1}-x_i)$. We derive, as special cases of our apparatus, known algorithms for the fused lasso and isotonic regression. Our approach also facilitates new variational penalties such as non-smooth barrier functions. We next derive and analyze multivariate problems in which $mathbf{x}_i,mathbf{y}_iinmathbb{R}^d$ and variational penalties that depend on $|mathbf{x}_{i+1}-mathbf{x}_i|$. The norms we consider are $ell_2$ and $ell_infty$ which promote group sparsity. Last but not least, we derive a lattice-based subgradient following for variational penalties characterized through the output of arbitrary convolutional filters. This paradigm yields efficient solvers for problems in which sparse high-order discrete derivatives such as acceleration and jerk are desirable. | [
"['Kai-Chia Mo' 'Shai Shalev-Shwartz' 'Nisæl Shártov']"
]
|
null | null | 2405.04714 | null | null | http://arxiv.org/pdf/2405.04714v1 | 2024-05-07T23:32:36Z | 2024-05-07T23:32:36Z | RACER: Epistemic Risk-Sensitive RL Enables Fast Driving with Fewer
Crashes | Reinforcement learning provides an appealing framework for robotic control due to its ability to learn expressive policies purely through real-world interaction. However, this requires addressing real-world constraints and avoiding catastrophic failures during training, which might severely impede both learning progress and the performance of the final policy. In many robotics settings, this amounts to avoiding certain "unsafe" states. The high-speed off-road driving task represents a particularly challenging instantiation of this problem: a high-return policy should drive as aggressively and as quickly as possible, which often requires getting close to the edge of the set of "safe" states, and therefore places a particular burden on the method to avoid frequent failures. To both learn highly performant policies and avoid excessive failures, we propose a reinforcement learning framework that combines risk-sensitive control with an adaptive action space curriculum. Furthermore, we show that our risk-sensitive objective automatically avoids out-of-distribution states when equipped with an estimator for epistemic uncertainty. We implement our algorithm on a small-scale rally car and show that it is capable of learning high-speed policies for a real-world off-road driving task. We show that our method greatly reduces the number of safety violations during the training process, and actually leads to higher-performance policies in both driving and non-driving simulation environments with similar challenges. | [
"['Kyle Stachowicz' 'Sergey Levine']"
]
|
null | null | 2405.04715 | null | null | http://arxiv.org/pdf/2405.04715v2 | 2024-06-30T21:37:56Z | 2024-05-07T23:37:40Z | Causality Pursuit from Heterogeneous Environments via Neural Adversarial
Invariance Learning | Pursuing causality from data is a fundamental problem in scientific discovery, treatment intervention, and transfer learning. This paper introduces a novel algorithmic method for addressing nonparametric invariance and causality learning in regression models across multiple environments, where the joint distribution of response variables and covariates varies, but the conditional expectations of outcome given an unknown set of quasi-causal variables are invariant. The challenge of finding such an unknown set of quasi-causal or invariant variables is compounded by the presence of endogenous variables that have heterogeneous effects across different environments, including even one of them in the regression would make the estimation inconsistent. The proposed Focused Adversial Invariant Regularization (FAIR) framework utilizes an innovative minimax optimization approach that breaks down the barriers, driving regression models toward prediction-invariant solutions through adversarial testing. Leveraging the representation power of neural networks, FAIR neural networks (FAIR-NN) are introduced for causality pursuit. It is shown that FAIR-NN can find the invariant variables and quasi-causal variables under a minimal identification condition and that the resulting procedure is adaptive to low-dimensional composition structures in a non-asymptotic analysis. Under a structural causal model, variables identified by FAIR-NN represent pragmatic causality and provably align with exact causal mechanisms under conditions of sufficient heterogeneity. Computationally, FAIR-NN employs a novel Gumbel approximation with decreased temperature and stochastic gradient descent ascent algorithm. The procedures are convincingly demonstrated using simulated and real-data examples. | [
"['Yihong Gu' 'Cong Fang' 'Peter Bühlmann' 'Jianqing Fan']"
]
|
null | null | 2405.04716 | null | null | http://arxiv.org/pdf/2405.04716v1 | 2024-05-07T23:43:46Z | 2024-05-07T23:43:46Z | Physics-based deep learning reveals rising heating demand heightens air
pollution in Norwegian cities | Policymakers frequently analyze air quality and climate change in isolation, disregarding their interactions. This study explores the influence of specific climate factors on air quality by contrasting a regression model with K-Means Clustering, Hierarchical Clustering, and Random Forest techniques. We employ Physics-based Deep Learning (PBDL) and Long Short-Term Memory (LSTM) to examine the air pollution predictions. Our analysis utilizes ten years (2009-2018) of daily traffic, weather, and air pollution data from three major cities in Norway. Findings from feature selection reveal a correlation between rising heating degree days and heightened air pollution levels, suggesting increased heating activities in Norway are a contributing factor to worsening air quality. PBDL demonstrates superior accuracy in air pollution predictions compared to LSTM. This paper contributes to the growing literature on PBDL methods for more accurate air pollution predictions using environmental variables, aiding policymakers in formulating effective data-driven climate policies. | [
"['Cong Cao' 'Ramit Debnath' 'R. Michael Alvarez']"
]
|
null | null | 2405.04746 | null | null | http://arxiv.org/pdf/2405.04746v1 | 2024-05-08T01:22:47Z | 2024-05-08T01:22:47Z | SVD-AE: Simple Autoencoders for Collaborative Filtering | Collaborative filtering (CF) methods for recommendation systems have been extensively researched, ranging from matrix factorization and autoencoder-based to graph filtering-based methods. Recently, lightweight methods that require almost no training have been recently proposed to reduce overall computation. However, existing methods still have room to improve the trade-offs among accuracy, efficiency, and robustness. In particular, there are no well-designed closed-form studies for emph{balanced} CF in terms of the aforementioned trade-offs. In this paper, we design SVD-AE, a simple yet effective singular vector decomposition (SVD)-based linear autoencoder, whose closed-form solution can be defined based on SVD for CF. SVD-AE does not require iterative training processes as its closed-form solution can be calculated at once. Furthermore, given the noisy nature of the rating matrix, we explore the robustness against such noisy interactions of existing CF methods and our SVD-AE. As a result, we demonstrate that our simple design choice based on truncated SVD can be used to strengthen the noise robustness of the recommendation while improving efficiency. Code is available at https://github.com/seoyoungh/svd-ae. | [
"['Seoyoung Hong' 'Jeongwhan Choi' 'Yeon-Chang Lee' 'Srijan Kumar'\n 'Noseong Park']"
]
|
null | null | 2405.04755 | null | null | http://arxiv.org/pdf/2405.04755v1 | 2024-05-08T01:51:19Z | 2024-05-08T01:51:19Z | Conditional Local Feature Encoding for Graph Neural Networks | Graph neural networks (GNNs) have shown great success in learning from graph-based data. The key mechanism of current GNNs is message passing, where a node's feature is updated based on the information passing from its local neighbourhood. A limitation of this mechanism is that node features become increasingly dominated by the information aggregated from the neighbourhood as we use more rounds of message passing. Consequently, as the GNN layers become deeper, adjacent node features tends to be similar, making it more difficult for GNNs to distinguish adjacent nodes, thereby, limiting the performance of GNNs. In this paper, we propose conditional local feature encoding (CLFE) to help prevent the problem of node features being dominated by the information from local neighbourhood. The idea of our method is to extract the node hidden state embedding from message passing process and concatenate it with the nodes feature from previous stage, then we utilise linear transformation to form a CLFE based on the concatenated vector. The CLFE will form the layer output to better preserve node-specific information, thus help to improve the performance of GNN models. To verify the feasibility of our method, we conducted extensive experiments on seven benchmark datasets for four graph domain tasks: super-pixel graph classification, node classification, link prediction, and graph regression. The experimental results consistently demonstrate that our method improves model performance across a variety of baseline GNN models for all four tasks. | [
"['Yongze Wang' 'Haimin Zhang' 'Qiang Wu' 'Min Xu']"
]
|
null | null | 2405.04756 | null | null | http://arxiv.org/pdf/2405.04756v1 | 2024-05-08T01:51:29Z | 2024-05-08T01:51:29Z | BiasKG: Adversarial Knowledge Graphs to Induce Bias in Large Language
Models | Modern large language models (LLMs) have a significant amount of world knowledge, which enables strong performance in commonsense reasoning and knowledge-intensive tasks when harnessed properly. The language model can also learn social biases, which has a significant potential for societal harm. There have been many mitigation strategies proposed for LLM safety, but it is unclear how effective they are for eliminating social biases. In this work, we propose a new methodology for attacking language models with knowledge graph augmented generation. We refactor natural language stereotypes into a knowledge graph, and use adversarial attacking strategies to induce biased responses from several open- and closed-source language models. We find our method increases bias in all models, even those trained with safety guardrails. This demonstrates the need for further research in AI safety, and further work in this new adversarial space. | [
"['Chu Fei Luo' 'Ahmad Ghawanmeh' 'Xiaodan Zhu' 'Faiza Khan Khattak']"
]
|
null | null | 2405.04759 | null | null | http://arxiv.org/pdf/2405.04759v2 | 2024-05-13T01:39:34Z | 2024-05-08T02:05:38Z | Multi-Label Out-of-Distribution Detection with Spectral Normalized Joint
Energy | In today's interconnected world, achieving reliable out-of-distribution (OOD) detection poses a significant challenge for machine learning models. While numerous studies have introduced improved approaches for multi-class OOD detection tasks, the investigation into multi-label OOD detection tasks has been notably limited. We introduce Spectral Normalized Joint Energy (SNoJoE), a method that consolidates label-specific information across multiple labels through the theoretically justified concept of an energy-based function. Throughout the training process, we employ spectral normalization to manage the model's feature space, thereby enhancing model efficacy and generalization, in addition to bolstering robustness. Our findings indicate that the application of spectral normalization to joint energy scores notably amplifies the model's capability for OOD detection. We perform OOD detection experiments utilizing PASCAL-VOC as the in-distribution dataset and ImageNet-22K or Texture as the out-of-distribution datasets. Our experimental results reveal that, in comparison to prior top performances, SNoJoE achieves 11% and 54% relative reductions in FPR95 on the respective OOD datasets, thereby defining the new state of the art in this field of study. | [
"['Yihan Mei' 'Xinyu Wang' 'Dell Zhang' 'Xiaoling Wang']"
]
|
null | null | 2405.04765 | null | null | http://arxiv.org/pdf/2405.04765v1 | 2024-05-08T02:24:09Z | 2024-05-08T02:24:09Z | When Foresight Pruning Meets Zeroth-Order Optimization: Efficient
Federated Learning for Low-Memory Devices | Although Federated Learning (FL) enables collaborative learning in Artificial Intelligence of Things (AIoT) design, it fails to work on low-memory AIoT devices due to its heavy memory usage. To address this problem, various federated pruning methods are proposed to reduce memory usage during inference. However, few of them can substantially mitigate the memory burdens during pruning and training. As an alternative, zeroth-order or backpropagation-free (BP-Free) methods can partially alleviate the memory consumption, but they suffer from scaling up and large computation overheads, since the gradient estimation error and floating point operations (FLOPs) increase as the dimensionality of the model parameters grows. In this paper, we propose a federated foresight pruning method based on Neural Tangent Kernel (NTK), which can seamlessly integrate with federated BP-Free training frameworks. We present an approximation to the computation of federated NTK by using the local NTK matrices. Moreover, we demonstrate that the data-free property of our method can substantially reduce the approximation error in extreme data heterogeneity scenarios. Since our approach improves the performance of the vanilla BP-Free method with fewer FLOPs and truly alleviates memory pressure during training and inference, it makes FL more friendly to low-memory devices. Comprehensive experimental results obtained from simulation- and real test-bed-based platforms show that our federated foresight-pruning method not only preserves the ability of the dense model with a memory reduction up to 9x but also boosts the performance of the vanilla BP-Free method with dramatically fewer FLOPs. | [
"['Pengyu Zhang' 'Yingjie Liu' 'Yingbo Zhou' 'Xiao Du' 'Xian Wei'\n 'Ting Wang' 'Mingsong Chen']"
]
|
null | null | 2405.04767 | null | null | http://arxiv.org/pdf/2405.04767v1 | 2024-05-08T02:31:51Z | 2024-05-08T02:31:51Z | Test-Time Augmentation for Traveling Salesperson Problem | We propose Test-Time Augmentation (TTA) as an effective technique for addressing combinatorial optimization problems, including the Traveling Salesperson Problem. In general, deep learning models possessing the property of invariance, where the output is uniquely determined regardless of the node indices, have been proposed to learn graph structures efficiently. In contrast, we interpret the permutation of node indices, which exchanges the elements of the distance matrix, as a TTA scheme. The results demonstrate that our method is capable of obtaining shorter solutions than the latest models. Furthermore, we show that the probability of finding a solution closer to an exact solution increases depending on the augmentation size. | [
"['Ryo Ishiyama' 'Takahiro Shirakawa' 'Seiichi Uchida' 'Shinnosuke Matsuo']"
]
|
null | null | 2405.04769 | null | null | http://arxiv.org/pdf/2405.04769v1 | 2024-05-08T02:33:35Z | 2024-05-08T02:33:35Z | Inference With Combining Rules From Multiple Differentially Private
Synthetic Datasets | Differential privacy (DP) has been accepted as a rigorous criterion for measuring the privacy protection offered by random mechanisms used to obtain statistics or, as we will study here, synthetic datasets from confidential data. Methods to generate such datasets are increasingly numerous, using varied tools including Bayesian models, deep neural networks and copulas. However, little is still known about how to properly perform statistical inference with these differentially private synthetic (DIPS) datasets. The challenge is for the analyses to take into account the variability from the synthetic data generation in addition to the usual sampling variability. A similar challenge also occurs when missing data is imputed before analysis, and statisticians have developed appropriate inference procedures for this case, which we tend extended to the case of synthetic datasets for privacy. In this work, we study the applicability of these procedures, based on combining rules, to the analysis of DIPS datasets. Our empirical experiments show that the proposed combining rules may offer accurate inference in certain contexts, but not in all cases. | [
"['Leila Nombo' 'Anne-Sophie Charest']"
]
|
null | null | 2405.04773 | null | null | http://arxiv.org/pdf/2405.04773v2 | 2024-05-28T09:19:55Z | 2024-05-08T02:44:13Z | Hypergraph-enhanced Dual Semi-supervised Graph Classification | In this paper, we study semi-supervised graph classification, which aims at accurately predicting the categories of graphs in scenarios with limited labeled graphs and abundant unlabeled graphs. Despite the promising capability of graph neural networks (GNNs), they typically require a large number of costly labeled graphs, while a wealth of unlabeled graphs fail to be effectively utilized. Moreover, GNNs are inherently limited to encoding local neighborhood information using message-passing mechanisms, thus lacking the ability to model higher-order dependencies among nodes. To tackle these challenges, we propose a Hypergraph-Enhanced DuAL framework named HEAL for semi-supervised graph classification, which captures graph semantics from the perspective of the hypergraph and the line graph, respectively. Specifically, to better explore the higher-order relationships among nodes, we design a hypergraph structure learning to adaptively learn complex node dependencies beyond pairwise relations. Meanwhile, based on the learned hypergraph, we introduce a line graph to capture the interaction between hyperedges, thereby better mining the underlying semantic structures. Finally, we develop a relational consistency learning to facilitate knowledge transfer between the two branches and provide better mutual guidance. Extensive experiments on real-world graph datasets verify the effectiveness of the proposed method against existing state-of-the-art methods. | [
"['Wei Ju' 'Zhengyang Mao' 'Siyu Yi' 'Yifang Qin' 'Yiyang Gu'\n 'Zhiping Xiao' 'Yifan Wang' 'Xiao Luo' 'Ming Zhang']"
]
|
null | null | 2405.04793 | null | null | http://arxiv.org/pdf/2405.04793v1 | 2024-05-08T03:57:45Z | 2024-05-08T03:57:45Z | Zero-shot LLM-guided Counterfactual Generation for Text | Counterfactual examples are frequently used for model development and evaluation in many natural language processing (NLP) tasks. Although methods for automated counterfactual generation have been explored, such methods depend on models such as pre-trained language models that are then fine-tuned on auxiliary, often task-specific datasets. Collecting and annotating such datasets for counterfactual generation is labor intensive and therefore, infeasible in practice. Therefore, in this work, we focus on a novel problem setting: textit{zero-shot counterfactual generation}. To this end, we propose a structured way to utilize large language models (LLMs) as general purpose counterfactual example generators. We hypothesize that the instruction-following and textual understanding capabilities of recent LLMs can be effectively leveraged for generating high quality counterfactuals in a zero-shot manner, without requiring any training or fine-tuning. Through comprehensive experiments on various downstream tasks in natural language processing (NLP), we demonstrate the efficacy of LLMs as zero-shot counterfactual generators in evaluating and explaining black-box NLP models. | [
"['Amrita Bhattacharjee' 'Raha Moraffah' 'Joshua Garland' 'Huan Liu']"
]
|
null | null | 2405.04795 | null | null | http://arxiv.org/pdf/2405.04795v3 | 2024-06-19T16:06:23Z | 2024-05-08T04:01:40Z | Variational Schrödinger Diffusion Models | Schr"odinger bridge (SB) has emerged as the go-to method for optimizing transportation plans in diffusion models. However, SB requires estimating the intractable forward score functions, inevitably resulting in the costly implicit training loss based on simulated trajectories. To improve the scalability while preserving efficient transportation plans, we leverage variational inference to linearize the forward score functions (variational scores) of SB and restore simulation-free properties in training backward scores. We propose the variational Schr"odinger diffusion model (VSDM), where the forward process is a multivariate diffusion and the variational scores are adaptively optimized for efficient transport. Theoretically, we use stochastic approximation to prove the convergence of the variational scores and show the convergence of the adaptively generated samples based on the optimal variational scores. Empirically, we test the algorithm in simulated examples and observe that VSDM is efficient in generations of anisotropic shapes and yields straighter sample trajectories compared to the single-variate diffusion. We also verify the scalability of the algorithm in real-world data and achieve competitive unconditional generation performance in CIFAR10 and conditional generation in time series modeling. Notably, VSDM no longer depends on warm-up initializations and has become tuning-friendly in training large-scale experiments. | [
"['Wei Deng' 'Weijian Luo' 'Yixin Tan' 'Marin Biloš' 'Yu Chen'\n 'Yuriy Nevmyvaka' 'Ricky T. Q. Chen']"
]
|
null | null | 2405.04800 | null | null | http://arxiv.org/pdf/2405.04800v1 | 2024-05-08T04:21:03Z | 2024-05-08T04:21:03Z | DeepDamageNet: A two-step deep-learning model for multi-disaster
building damage segmentation and classification using satellite imagery | Satellite imagery has played an increasingly important role in post-disaster building damage assessment. Unfortunately, current methods still rely on manual visual interpretation, which is often time-consuming and can cause very low accuracy. To address the limitations of manual interpretation, there has been a significant increase in efforts to automate the process. We present a solution that performs the two most important tasks in building damage assessment, segmentation and classification, through deep-learning models. We show our results submitted as part of the xView2 Challenge, a competition to design better models for identifying buildings and their damage level after exposure to multiple kinds of natural disasters. Our best model couples a building identification semantic segmentation convolutional neural network (CNN) to a building damage classification CNN, with a combined F1 score of 0.66, surpassing the xView2 challenge baseline F1 score of 0.28. We find that though our model was able to identify buildings with relatively high accuracy, building damage classification across various disaster types is a difficult task due to the visual similarity between different damage levels and different damage distribution between disaster types, highlighting the fact that it may be important to have a probabilistic prior estimate regarding disaster damage in order to obtain accurate predictions. | [
"['Irene Alisjahbana' 'Jiawei Li' 'Ben' 'Strong' 'Yue Zhang']"
]
|
null | null | 2405.04815 | null | null | http://arxiv.org/pdf/2405.04815v1 | 2024-05-08T05:29:38Z | 2024-05-08T05:29:38Z | Proportion Estimation by Masked Learning from Label Proportion | The PD-L1 rate, the number of PD-L1 positive tumor cells over the total number of all tumor cells, is an important metric for immunotherapy. This metric is recorded as diagnostic information with pathological images. In this paper, we propose a proportion estimation method with a small amount of cell-level annotation and proportion annotation, which can be easily collected. Since the PD-L1 rate is calculated from only `tumor cells' and not using `non-tumor cells', we first detect tumor cells with a detection model. Then, we estimate the PD-L1 proportion by introducing a masking technique to `learning from label proportion.' In addition, we propose a weighted focal proportion loss to address data imbalance problems. Experiments using clinical data demonstrate the effectiveness of our method. Our method achieved the best performance in comparisons. | [
"['Takumi Okuo' 'Kazuya Nishimura' 'Hiroaki Ito' 'Kazuhiro Terada'\n 'Akihiko Yoshizawa' 'Ryoma Bise']"
]
|
null | null | 2405.04825 | null | null | http://arxiv.org/pdf/2405.04825v1 | 2024-05-08T05:49:46Z | 2024-05-08T05:49:46Z | Explanation as a Watermark: Towards Harmless and Multi-bit Model
Ownership Verification via Watermarking Feature Attribution | Ownership verification is currently the most critical and widely adopted post-hoc method to safeguard model copyright. In general, model owners exploit it to identify whether a given suspicious third-party model is stolen from them by examining whether it has particular properties `inherited' from their released models. Currently, backdoor-based model watermarks are the primary and cutting-edge methods to implant such properties in the released models. However, backdoor-based methods have two fatal drawbacks, including harmfulness and ambiguity. The former indicates that they introduce maliciously controllable misclassification behaviors ($i.e.$, backdoor) to the watermarked released models. The latter denotes that malicious users can easily pass the verification by finding other misclassified samples, leading to ownership ambiguity. In this paper, we argue that both limitations stem from the `zero-bit' nature of existing watermarking schemes, where they exploit the status ($i.e.$, misclassified) of predictions for verification. Motivated by this understanding, we design a new watermarking paradigm, $i.e.$, Explanation as a Watermark (EaaW), that implants verification behaviors into the explanation of feature attribution instead of model predictions. Specifically, EaaW embeds a `multi-bit' watermark into the feature attribution explanation of specific trigger samples without changing the original prediction. We correspondingly design the watermark embedding and extraction algorithms inspired by explainable artificial intelligence. In particular, our approach can be used for different tasks ($e.g.$, image classification and text generation). Extensive experiments verify the effectiveness and harmlessness of our EaaW and its resistance to potential attacks. | [
"['Shuo Shao' 'Yiming Li' 'Hongwei Yao' 'Yiling He' 'Zhan Qin' 'Kui Ren']"
]
|
null | null | 2405.04841 | null | null | http://arxiv.org/pdf/2405.04841v1 | 2024-05-08T06:29:26Z | 2024-05-08T06:29:26Z | xMTrans: Temporal Attentive Cross-Modality Fusion Transformer for
Long-Term Traffic Prediction | Traffic predictions play a crucial role in intelligent transportation systems. The rapid development of IoT devices allows us to collect different kinds of data with high correlations to traffic predictions, fostering the development of efficient multi-modal traffic prediction models. Until now, there are few studies focusing on utilizing advantages of multi-modal data for traffic predictions. In this paper, we introduce a novel temporal attentive cross-modality transformer model for long-term traffic predictions, namely xMTrans, with capability of exploring the temporal correlations between the data of two modalities: one target modality (for prediction, e.g., traffic congestion) and one support modality (e.g., people flow). We conducted extensive experiments to evaluate our proposed model on traffic congestion and taxi demand predictions using real-world datasets. The results showed the superiority of xMTrans against recent state-of-the-art methods on long-term traffic predictions. In addition, we also conducted a comprehensive ablation study to further analyze the effectiveness of each module in xMTrans. | [
"['Huy Quang Ung' 'Hao Niu' 'Minh-Son Dao' 'Shinya Wada'\n 'Atsunori Minamikawa']"
]
|
null | null | 2405.04854 | null | null | http://arxiv.org/pdf/2405.04854v1 | 2024-05-08T07:09:43Z | 2024-05-08T07:09:43Z | Explaining Clustering of Ecological Momentary Assessment Data Through
Temporal and Feature Attention | In the field of psychopathology, Ecological Momentary Assessment (EMA) studies offer rich individual data on psychopathology-relevant variables (e.g., affect, behavior, etc) in real-time. EMA data is collected dynamically, represented as complex multivariate time series (MTS). Such information is crucial for a better understanding of mental disorders at the individual- and group-level. More specifically, clustering individuals in EMA data facilitates uncovering and studying the commonalities as well as variations of groups in the population. Nevertheless, since clustering is an unsupervised task and true EMA grouping is not commonly available, the evaluation of clustering is quite challenging. An important aspect of evaluation is clustering explainability. Thus, this paper proposes an attention-based interpretable framework to identify the important time-points and variables that play primary roles in distinguishing between clusters. A key part of this study is to examine ways to analyze, summarize, and interpret the attention weights as well as evaluate the patterns underlying the important segments of the data that differentiate across clusters. To evaluate the proposed approach, an EMA dataset of 187 individuals grouped in 3 clusters is used for analyzing the derived attention-based importance attributes. More specifically, this analysis provides the distinct characteristics at the cluster-, feature- and individual level. Such clustering explanations could be beneficial for generalizing existing concepts of mental disorders, discovering new insights, and even enhancing our knowledge at an individual level. | [
"['Mandani Ntekouli' 'Gerasimos Spanakis' 'Lourens Waldorp' 'Anne Roefs']"
]
|
null | null | 2405.04865 | null | null | http://arxiv.org/pdf/2405.04865v3 | 2024-06-12T10:05:11Z | 2024-05-08T07:43:43Z | Regime Learning for Differentiable Particle Filters | Differentiable particle filters are an emerging class of models that combine sequential Monte Carlo techniques with the flexibility of neural networks to perform state space inference. This paper concerns the case where the system may switch between a finite set of state-space models, i.e. regimes. No prior approaches effectively learn both the individual regimes and the switching process simultaneously. In this paper, we propose the neural network based regime learning differentiable particle filter (RLPF) to address this problem. We further design a training procedure for the RLPF and other related algorithms. We demonstrate competitive performance compared to the previous state-of-the-art algorithms on a pair of numerical experiments. | [
"['John-Joseph Brady' 'Yuhui Luo' 'Wenwu Wang' 'Victor Elvira' 'Yunpeng Li']"
]
|
null | null | 2405.04875 | null | null | http://arxiv.org/pdf/2405.04875v1 | 2024-05-08T08:12:21Z | 2024-05-08T08:12:21Z | SCALA: Split Federated Learning with Concatenated Activations and Logit
Adjustments | Split Federated Learning (SFL) is a distributed machine learning framework which strategically divides the learning process between a server and clients and collaboratively trains a shared model by aggregating local models updated based on data from distributed clients. However, data heterogeneity and partial client participation result in label distribution skew, which severely degrades the learning performance. To address this issue, we propose SFL with Concatenated Activations and Logit Adjustments (SCALA). Specifically, the activations from the client-side models are concatenated as the input of the server-side model so as to centrally adjust label distribution across different clients, and logit adjustments of loss functions on both server-side and client-side models are performed to deal with the label distribution variation across different subsets of participating clients. Theoretical analysis and experimental results verify the superiority of the proposed SCALA on public datasets. | [
"['Jiarong Yang' 'Yuan Liu']"
]
|
null | null | 2405.04881 | null | null | http://arxiv.org/pdf/2405.04881v1 | 2024-05-08T08:30:34Z | 2024-05-08T08:30:34Z | Gödel Number based Clustering Algorithm with Decimal First Degree
Cellular Automata | In this paper, a decimal first degree cellular automata (FDCA) based clustering algorithm is proposed where clusters are created based on reachability. Cyclic spaces are created and configurations which are in the same cycle are treated as the same cluster. Here, real-life data objects are encoded into decimal strings using G"odel number based encoding. The benefits of the scheme is, it reduces the encoded string length while maintaining the features properties. Candidate CA rules are identified based on some theoretical criteria such as self-replication and information flow. An iterative algorithm is developed to generate the desired number of clusters over three stages. The results of the clustering are evaluated based on benchmark clustering metrics such as Silhouette score, Davis Bouldin, Calinski Harabasz and Dunn Index. In comparison with the existing state-of-the-art clustering algorithms, our proposed algorithm gives better performance. | [
"['Vicky Vikrant' 'Narodia Parth P' 'Kamalika Bhattacharjee']"
]
|
null | null | 2405.04883 | null | null | http://arxiv.org/pdf/2405.04883v2 | 2024-05-10T07:18:00Z | 2024-05-08T08:32:34Z | FreeBind: Free Lunch in Unified Multimodal Space via Knowledge Fusion | Unified multi-model representation spaces are the foundation of multimodal understanding and generation. However, the billions of model parameters and catastrophic forgetting problems make it challenging to further enhance pre-trained unified spaces. In this work, we propose FreeBind, an idea that treats multimodal representation spaces as basic units, and freely augments pre-trained unified space by integrating knowledge from extra expert spaces via "space bonds". Specifically, we introduce two kinds of basic space bonds: 1) Space Displacement Bond and 2) Space Combination Bond. Based on these basic bonds, we design Complex Sequential & Parallel Bonds to effectively integrate multiple spaces simultaneously. Benefiting from the modularization concept, we further propose a coarse-to-fine customized inference strategy to flexibly adjust the enhanced unified space for different purposes. Experimentally, we bind ImageBind with extra image-text and audio-text expert spaces, resulting in three main variants: ImageBind++, InternVL_IB, and InternVL_IB++. These resulting spaces outperform ImageBind on 5 audio-image-text downstream tasks across 9 datasets. Moreover, via customized inference, it even surpasses the advanced audio-text and image-text expert spaces. | [
"['Zehan Wang' 'Ziang Zhang' 'Xize Cheng' 'Rongjie Huang' 'Luping Liu'\n 'Zhenhui Ye' 'Haifeng Huang' 'Yang Zhao' 'Tao Jin' 'Peng Gao' 'Zhou Zhao']"
]
|
null | null | 2405.04903 | null | null | http://arxiv.org/pdf/2405.04903v2 | 2024-05-17T08:32:12Z | 2024-05-08T09:16:54Z | Imbalanced Graph Classification with Multi-scale Oversampling Graph
Neural Networks | One main challenge in imbalanced graph classification is to learn expressive representations of the graphs in under-represented (minority) classes. Existing generic imbalanced learning methods, such as oversampling and imbalanced learning loss functions, can be adopted for enabling graph representation learning models to cope with this challenge. However, these methods often directly operate on the graph representations, ignoring rich discriminative information within the graphs and their interactions. To tackle this issue, we introduce a novel multi-scale oversampling graph neural network (MOSGNN) that learns expressive minority graph representations based on intra- and inter-graph semantics resulting from oversampled graphs at multiple scales - subgraph, graph, and pairwise graphs. It achieves this by jointly optimizing subgraph-level, graph-level, and pairwise-graph learning tasks to learn the discriminative information embedded within and between the minority graphs. Extensive experiments on 16 imbalanced graph datasets show that MOSGNN i) significantly outperforms five state-of-the-art models, and ii) offers a generic framework, in which different advanced imbalanced learning loss functions can be easily plugged in and obtain significantly improved classification performance. | [
"['Rongrong Ma' 'Guansong Pang' 'Ling Chen']"
]
|
null | null | 2405.04910 | null | null | http://arxiv.org/pdf/2405.04910v1 | 2024-05-08T09:28:26Z | 2024-05-08T09:28:26Z | Learning with Posterior Sampling for Revenue Management under
Time-varying Demand | This paper discusses the revenue management (RM) problem to maximize revenue by pricing items or services. One challenge in this problem is that the demand distribution is unknown and varies over time in real applications such as airline and retail industries. In particular, the time-varying demand has not been well studied under scenarios of unknown demand due to the difficulty of jointly managing the remaining inventory and estimating the demand. To tackle this challenge, we first introduce an episodic generalization of the RM problem motivated by typical application scenarios. We then propose a computationally efficient algorithm based on posterior sampling, which effectively optimizes prices by solving linear programming. We derive a Bayesian regret upper bound of this algorithm for general models where demand parameters can be correlated between time periods, while also deriving a regret lower bound for generic algorithms. Our empirical study shows that the proposed algorithm performs better than other benchmark algorithms and comparably to the optimal policy in hindsight. We also propose a heuristic modification of the proposed algorithm, which further efficiently learns the pricing policy in the experiments. | [
"['Kazuma Shimizu' 'Junya Honda' 'Shinji Ito' 'Shinji Nakadai']"
]
|
null | null | 2405.04912 | null | null | http://arxiv.org/pdf/2405.04912v1 | 2024-04-04T16:20:06Z | 2024-04-04T16:20:06Z | GP-MoLFormer: A Foundation Model For Molecular Generation | Transformer-based models trained on large and general purpose datasets consisting of molecular strings have recently emerged as a powerful tool for successfully modeling various structure-property relations. Inspired by this success, we extend the paradigm of training chemical language transformers on large-scale chemical datasets to generative tasks in this work. Specifically, we propose GP-MoLFormer, an autoregressive molecular string generator that is trained on more than 1.1B chemical SMILES. GP-MoLFormer uses a 46.8M parameter transformer decoder model with linear attention and rotary positional encodings as the base architecture. We explore the utility of GP-MoLFormer in generating novel, valid, and unique SMILES. Impressively, we find GP-MoLFormer is able to generate a significant fraction of novel, valid, and unique SMILES even when the number of generated molecules is in the 10 billion range and the reference set is over a billion. We also find strong memorization of training data in GP-MoLFormer generations, which has so far remained unexplored for chemical language models. Our analyses reveal that training data memorization and novelty in generations are impacted by the quality of the training data; duplication bias in training data can enhance memorization at the cost of lowering novelty. We evaluate GP-MoLFormer's utility and compare it with that of existing baselines on three different tasks: de novo generation, scaffold-constrained molecular decoration, and unconstrained property-guided optimization. While the first two are handled with no additional training, we propose a parameter-efficient fine-tuning method for the last task, which uses property-ordered molecular pairs as input. We call this new approach pair-tuning. Our results show GP-MoLFormer performs better or comparable with baselines across all three tasks, demonstrating its general utility. | [
"['Jerret Ross' 'Brian Belgodere' 'Samuel C. Hoffman'\n 'Vijil Chenthamarakshan' 'Youssef Mroueh' 'Payel Das']"
]
|
null | null | 2405.04919 | null | null | http://arxiv.org/pdf/2405.04919v1 | 2024-05-08T09:41:25Z | 2024-05-08T09:41:25Z | Fast Computation of Leave-One-Out Cross-Validation for $k$-NN Regression | We describe a fast computation method for leave-one-out cross-validation (LOOCV) for $k$-nearest neighbours ($k$-NN) regression. We show that, under a tie-breaking condition for nearest neighbours, the LOOCV estimate of the mean square error for $k$-NN regression is identical to the mean square error of $(k+1)$-NN regression evaluated on the training data, multiplied by the scaling factor $(k+1)^2/k^2$. Therefore, to compute the LOOCV score, one only needs to fit $(k+1)$-NN regression only once, and does not need to repeat training-validation of $k$-NN regression for the number of training data. Numerical experiments confirm the validity of the fast computation method. | [
"['Motonobu Kanagawa']"
]
|
null | null | 2405.04923 | null | null | http://arxiv.org/pdf/2405.04923v2 | 2024-05-30T12:04:17Z | 2024-05-08T09:45:54Z | DataSP: A Differential All-to-All Shortest Path Algorithm for Learning
Costs and Predicting Paths with Context | Learning latent costs of transitions on graphs from trajectories demonstrations under various contextual features is challenging but useful for path planning. Yet, existing methods either oversimplify cost assumptions or scale poorly with the number of observed trajectories. This paper introduces DataSP, a differentiable all-to-all shortest path algorithm to facilitate learning latent costs from trajectories. It allows to learn from a large number of trajectories in each learning step without additional computation. Complex latent cost functions from contextual features can be represented in the algorithm through a neural network approximation. We further propose a method to sample paths from DataSP in order to reconstruct/mimic observed paths' distributions. We prove that the inferred distribution follows the maximum entropy principle. We show that DataSP outperforms state-of-the-art differentiable combinatorial solver and classical machine learning approaches in predicting paths on graphs. | [
"['Alan A. Lahoud' 'Erik Schaffernicht' 'Johannes A. Stork']"
]
|
null | null | 2405.04938 | null | null | http://arxiv.org/pdf/2405.04938v1 | 2024-05-08T10:10:24Z | 2024-05-08T10:10:24Z | Fault Identification Enhancement with Reinforcement Learning (FIERL) | This letter presents a novel approach in the field of Active Fault Detection (AFD), by explicitly separating the task into two parts: Passive Fault Detection (PFD) and control input design. This formulation is very general, and most existing AFD literature can be viewed through this lens. By recognizing this separation, PFD methods can be leveraged to provide components that make efficient use of the available information, while the control input is designed in order to optimize the gathering of information. The core contribution of this work is FIERL, a general simulation-based approach for the design of such control strategies, using Constrained Reinforcement Learning (CRL) to optimize the performance of arbitrary passive detectors. The control policy is learned without the need of knowing the passive detector inner workings, making FIERL broadly applicable. However, it is especially useful when paired with the design of an efficient passive component. Unlike most AFD approaches, FIERL can handle fairly complex scenarios such as continuous sets of fault modes. The effectiveness of FIERL is tested on a benchmark problem for actuator fault diagnosis, where FIERL is shown to be fairly robust, being able to generalize to fault dynamics not seen in training. | [
"['Valentina Zaccaria' 'Davide Sartor' 'Simone Del Favero'\n 'Gian Antonio Susto']"
]
|
null | null | 2405.04944 | null | null | http://arxiv.org/pdf/2405.04944v1 | 2024-05-08T10:28:20Z | 2024-05-08T10:28:20Z | A Sparse Tensor Generator with Efficient Feature Extraction | Sparse tensor operations are gaining attention in emerging applications such as social networks, deep learning, diagnosis, crime, and review analysis. However, a major obstacle for research in sparse tensor operations is the deficiency of a broad-scale sparse tensor dataset. Another challenge in sparse tensor operations is examining the sparse tensor features, which are not only important for revealing its nonzero pattern but also have a significant impact on determining the best-suited storage format, the decomposition algorithm, and the reordering methods. However, due to the large sizes of real tensors, even extracting these features becomes costly without caution. To address these gaps in the literature, we have developed a smart sparse tensor generator that mimics the substantial features of real sparse tensors. Moreover, we propose various methods for efficiently extracting an extensive set of features for sparse tensors. The effectiveness of our generator is validated through the quality of features and the performance of decomposition in the generated tensors. Both the sparse tensor feature extractor and the tensor generator are open source with all the artifacts available at https://github.com/sparcityeu/feaTen and https://github.com/sparcityeu/genTen, respectively. | [
"['Tugba Torun' 'Eren Yenigul' 'Ameer Taweel' 'Didem Unat']"
]
|
null | null | 2405.04953 | null | null | http://arxiv.org/pdf/2405.04953v2 | 2024-05-11T11:39:20Z | 2024-05-08T10:47:28Z | Supervised Anomaly Detection for Complex Industrial Images | Automating visual inspection in industrial production lines is essential for increasing product quality across various industries. Anomaly detection (AD) methods serve as robust tools for this purpose. However, existing public datasets primarily consist of images without anomalies, limiting the practical application of AD methods in production settings. To address this challenge, we present (1) the Valeo Anomaly Dataset (VAD), a novel real-world industrial dataset comprising 5000 images, including 2000 instances of challenging real defects across more than 20 subclasses. Acknowledging that traditional AD methods struggle with this dataset, we introduce (2) Segmentation-based Anomaly Detector (SegAD). First, SegAD leverages anomaly maps as well as segmentation maps to compute local statistics. Next, SegAD uses these statistics and an optional supervised classifier score as input features for a Boosted Random Forest (BRF) classifier, yielding the final anomaly score. Our SegAD achieves state-of-the-art performance on both VAD (+2.1% AUROC) and the VisA dataset (+0.4% AUROC). The code and the models are publicly available. | [
"['Aimira Baitieva' 'David Hurych' 'Victor Besnier' 'Olivier Bernard']"
]
|
null | null | 2405.04972 | null | null | http://arxiv.org/pdf/2405.04972v1 | 2024-05-08T11:25:04Z | 2024-05-08T11:25:04Z | Overcoming Anchoring Bias: The Potential of AI and XAI-based Decision
Support | Information systems (IS) are frequently designed to leverage the negative effect of anchoring bias to influence individuals' decision-making (e.g., by manipulating purchase decisions). Recent advances in Artificial Intelligence (AI) and the explanations of its decisions through explainable AI (XAI) have opened new opportunities for mitigating biased decisions. So far, the potential of these technological advances to overcome anchoring bias remains widely unclear. To this end, we conducted two online experiments with a total of N=390 participants in the context of purchase decisions to examine the impact of AI and XAI-based decision support on anchoring bias. Our results show that AI alone and its combination with XAI help to mitigate the negative effect of anchoring bias. Ultimately, our findings have implications for the design of AI and XAI-based decision support and IS to overcome cognitive biases. | [
"['Felix Haag' 'Carlo Stingl' 'Katrin Zerfass' 'Konstantin Hopf'\n 'Thorsten Staake']"
]
|
null | null | 2405.04984 | null | null | http://arxiv.org/pdf/2405.04984v1 | 2024-05-08T11:46:00Z | 2024-05-08T11:46:00Z | Dynamic Data Layout Optimization with Worst-case Guarantees | Many data analytics systems store and process large datasets in partitions containing millions of rows. By mapping rows to partitions in an optimized way, it is possible to improve query performance by skipping over large numbers of irrelevant partitions during query processing. This mapping is referred to as a data layout. Recent works have shown that customizing the data layout to the anticipated query workload greatly improves query performance, but the performance benefits may disappear if the workload changes. Reorganizing data layouts to accommodate workload drift can resolve this issue, but reorganization costs could exceed query savings if not done carefully. In this paper, we present an algorithmic framework OReO that makes online reorganization decisions to balance the benefits of improved query performance with the costs of reorganization. Our framework extends results from Metrical Task Systems to provide a tight bound on the worst-case performance guarantee for online reorganization, without prior knowledge of the query workload. Through evaluation on real-world datasets and query workloads, our experiments demonstrate that online reorganization with OReO can lead to an up to 32% improvement in combined query and reorganization time compared to using a single, optimized data layout for the entire workload. | [
"['Kexin Rong' 'Paul Liu' 'Sarah Ashok Sonje' 'Moses Charikar']"
]
|
null | null | 2405.04990 | null | null | http://arxiv.org/pdf/2405.04990v1 | 2024-05-08T11:54:15Z | 2024-05-08T11:54:15Z | Health Index Estimation Through Integration of General Knowledge with
Unsupervised Learning | Accurately estimating a Health Index (HI) from condition monitoring data (CM) is essential for reliable and interpretable prognostics and health management (PHM) in complex systems. In most scenarios, complex systems operate under varying operating conditions and can exhibit different fault modes, making unsupervised inference of an HI from CM data a significant challenge. Hybrid models combining prior knowledge about degradation with deep learning models have been proposed to overcome this challenge. However, previously suggested hybrid models for HI estimation usually rely heavily on system-specific information, limiting their transferability to other systems. In this work, we propose an unsupervised hybrid method for HI estimation that integrates general knowledge about degradation into the convolutional autoencoder's model architecture and learning algorithm, enhancing its applicability across various systems. The effectiveness of the proposed method is demonstrated in two case studies from different domains: turbofan engines and lithium batteries. The results show that the proposed method outperforms other competitive alternatives, including residual-based methods, in terms of HI quality and their utility for Remaining Useful Life (RUL) predictions. The case studies also highlight the comparable performance of our proposed method with a supervised model trained with HI labels. | [
"['Kristupas Bajarunas' 'Marcia L. Baptista' 'Kai Goebel' 'Manuel A. Chao']"
]
|
null | null | 2405.05015 | null | null | http://arxiv.org/pdf/2405.05015v1 | 2024-05-08T12:31:35Z | 2024-05-08T12:31:35Z | Concrete Dense Network for Long-Sequence Time Series Clustering | Time series clustering is fundamental in data analysis for discovering temporal patterns. Despite recent advancements, learning cluster-friendly representations is still challenging, particularly with long and complex time series. Deep temporal clustering methods have been trying to integrate the canonical k-means into end-to-end training of neural networks but fall back on surrogate losses due to the non-differentiability of the hard cluster assignment, yielding sub-optimal solutions. In addition, the autoregressive strategy used in the state-of-the-art RNNs is subject to error accumulation and slow training, while recent research findings have revealed that Transformers are less effective due to time points lacking semantic meaning, to the permutation invariance of attention that discards the chronological order and high computation cost. In light of these observations, we present LoSTer which is a novel dense autoencoder architecture for the long-sequence time series clustering problem (LSTC) capable of optimizing the k-means objective via the Gumbel-softmax reparameterization trick and designed specifically for accurate and fast clustering of long time series. Extensive experiments on numerous benchmark datasets and two real-world applications prove the effectiveness of LoSTer over state-of-the-art RNNs and Transformer-based deep clustering methods. | [
"['Redemptor Jr Laceda Taloma' 'Patrizio Pisani' 'Danilo Comminiello']"
]
|
null | null | 2405.05025 | null | null | http://arxiv.org/pdf/2405.05025v1 | 2024-05-08T12:56:33Z | 2024-05-08T12:56:33Z | Learning Structural Causal Models through Deep Generative Models:
Methods, Guarantees, and Challenges | This paper provides a comprehensive review of deep structural causal models (DSCMs), particularly focusing on their ability to answer counterfactual queries using observational data within known causal structures. It delves into the characteristics of DSCMs by analyzing the hypotheses, guarantees, and applications inherent to the underlying deep learning components and structural causal models, fostering a finer understanding of their capabilities and limitations in addressing different counterfactual queries. Furthermore, it highlights the challenges and open questions in the field of deep structural causal modeling. It sets the stages for researchers to identify future work directions and for practitioners to get an overview in order to find out the most appropriate methods for their needs. | [
"['Audrey Poinsot' 'Alessandro Leite' 'Nicolas Chesneau' 'Michèle Sébag'\n 'Marc Schoenauer']"
]
|
null | null | 2405.05033 | null | null | http://arxiv.org/pdf/2405.05033v1 | 2024-05-08T13:03:55Z | 2024-05-08T13:03:55Z | Multi-fidelity Hamiltonian Monte Carlo | Numerous applications in biology, statistics, science, and engineering require generating samples from high-dimensional probability distributions. In recent years, the Hamiltonian Monte Carlo (HMC) method has emerged as a state-of-the-art Markov chain Monte Carlo technique, exploiting the shape of such high-dimensional target distributions to efficiently generate samples. Despite its impressive empirical success and increasing popularity, its wide-scale adoption remains limited due to the high computational cost of gradient calculation. Moreover, applying this method is impossible when the gradient of the posterior cannot be computed (for example, with black-box simulators). To overcome these challenges, we propose a novel two-stage Hamiltonian Monte Carlo algorithm with a surrogate model. In this multi-fidelity algorithm, the acceptance probability is computed in the first stage via a standard HMC proposal using an inexpensive differentiable surrogate model, and if the proposal is accepted, the posterior is evaluated in the second stage using the high-fidelity (HF) numerical solver. Splitting the standard HMC algorithm into these two stages allows for approximating the gradient of the posterior efficiently, while producing accurate posterior samples by using HF numerical solvers in the second stage. We demonstrate the effectiveness of this algorithm for a range of problems, including linear and nonlinear Bayesian inverse problems with in-silico data and experimental data. The proposed algorithm is shown to seamlessly integrate with various low-fidelity and HF models, priors, and datasets. Remarkably, our proposed method outperforms the traditional HMC algorithm in both computational and statistical efficiency by several orders of magnitude, all while retaining or improving the accuracy in computed posterior statistics. | [
"['Dhruv V. Patel' 'Jonghyun Lee' 'Matthew W. Farthing'\n 'Peter K. Kitanidis' 'Eric F. Darve']"
]
|
null | null | 2405.05066 | null | null | http://arxiv.org/pdf/2405.05066v1 | 2024-05-08T14:04:35Z | 2024-05-08T14:04:35Z | Designing Skill-Compatible AI: Methodologies and Frameworks in Chess | Powerful artificial intelligence systems are often used in settings where they must interact with agents that are computationally much weaker, for example when they work alongside humans or operate in complex environments where some tasks are handled by algorithms, heuristics, or other entities of varying computational power. For AI agents to successfully interact in these settings, however, achieving superhuman performance alone is not sufficient; they also need to account for suboptimal actions or idiosyncratic style from their less-skilled counterparts. We propose a formal evaluation framework for assessing the compatibility of near-optimal AI with interaction partners who may have much lower levels of skill; we use popular collaborative chess variants as model systems to study and develop AI agents that can successfully interact with lower-skill entities. Traditional chess engines designed to output near-optimal moves prove to be inadequate partners when paired with engines of various lower skill levels in this domain, as they are not designed to consider the presence of other agents. We contribute three methodologies to explicitly create skill-compatible AI agents in complex decision-making settings, and two chess game frameworks designed to foster collaboration between powerful AI agents and less-skilled partners. On these frameworks, our agents outperform state-of-the-art chess AI (based on AlphaZero) despite being weaker in conventional chess, demonstrating that skill-compatibility is a tangible trait that is qualitatively and measurably distinct from raw performance. Our evaluations further explore and clarify the mechanisms by which our agents achieve skill-compatibility. | [
"['Karim Hamade' 'Reid McIlroy-Young' 'Siddhartha Sen' 'Jon Kleinberg'\n 'Ashton Anderson']"
]
|
null | null | 2405.05072 | null | null | http://arxiv.org/pdf/2405.05072v1 | 2024-05-08T14:14:03Z | 2024-05-08T14:14:03Z | Novel Actor-Critic Algorithm for Robust Decision Making of CAV under
Delays and Loss of V2X Data | Current autonomous driving systems heavily rely on V2X communication data to enhance situational awareness and the cooperation between vehicles. However, a major challenge when using V2X data is that it may not be available periodically because of unpredictable delays and data loss during wireless transmission between road stations and the receiver vehicle. This issue should be considered when designing control strategies for connected and autonomous vehicles. Therefore, this paper proposes a novel 'Blind Actor-Critic' algorithm that guarantees robust driving performance in V2X environment with delayed and/or lost data. The novel algorithm incorporates three key mechanisms: a virtual fixed sampling period, a combination of Temporal-Difference and Monte Carlo learning, and a numerical approximation of immediate reward values. To address the temporal aperiodicity problem of V2X data, we first illustrate this challenge. Then, we provide a detailed explanation of the Blind Actor-Critic algorithm where we highlight the proposed components to compensate for the temporal aperiodicity problem of V2X data. We evaluate the performance of our algorithm in a simulation environment and compare it to benchmark approaches. The results demonstrate that training metrics are improved compared to conventional actor-critic algorithms. Additionally, testing results show that our approach provides robust control, even under low V2X network reliability levels. | [
"['Zine el abidine Kherroubi']"
]
|
null | null | 2405.05075 | null | null | http://arxiv.org/pdf/2405.05075v1 | 2024-05-08T14:18:13Z | 2024-05-08T14:18:13Z | Towards Efficient Training and Evaluation of Robust Models against $l_0$
Bounded Adversarial Perturbations | This work studies sparse adversarial perturbations bounded by $l_0$ norm. We propose a white-box PGD-like attack method named sparse-PGD to effectively and efficiently generate such perturbations. Furthermore, we combine sparse-PGD with a black-box attack to comprehensively and more reliably evaluate the models' robustness against $l_0$ bounded adversarial perturbations. Moreover, the efficiency of sparse-PGD enables us to conduct adversarial training to build robust models against sparse perturbations. Extensive experiments demonstrate that our proposed attack algorithm exhibits strong performance in different scenarios. More importantly, compared with other robust models, our adversarially trained model demonstrates state-of-the-art robustness against various sparse attacks. Codes are available at https://github.com/CityU-MLO/sPGD. | [
"['Xuyang Zhong' 'Yixiao Huang' 'Chen Liu']"
]
|
null | null | 2405.05081 | null | null | http://arxiv.org/pdf/2405.05081v1 | 2024-05-08T14:25:40Z | 2024-05-08T14:25:40Z | Robust deep learning from weakly dependent data | Recent developments on deep learning established some theoretical properties of deep neural networks estimators. However, most of the existing works on this topic are restricted to bounded loss functions or (sub)-Gaussian or bounded input. This paper considers robust deep learning from weakly dependent observations, with unbounded loss function and unbounded input/output. It is only assumed that the output variable has a finite $r$ order moment, with $r >1$. Non asymptotic bounds for the expected excess risk of the deep neural network estimator are established under strong mixing, and $psi$-weak dependence assumptions on the observations. We derive a relationship between these bounds and $r$, and when the data have moments of any order (that is $r=infty$), the convergence rate is close to some well-known results. When the target predictor belongs to the class of H"older smooth functions with sufficiently large smoothness index, the rate of the expected excess risk for exponentially strongly mixing data is close to or as same as those for obtained with i.i.d. samples. Application to robust nonparametric regression and robust nonparametric autoregression are considered. The simulation study for models with heavy-tailed errors shows that, robust estimators with absolute loss and Huber loss function outperform the least squares method. | [
"['William Kengne' 'Modou Wade']"
]
|
null | null | 2405.05097 | null | null | http://arxiv.org/pdf/2405.05097v3 | 2024-07-01T13:46:06Z | 2024-05-08T14:49:27Z | Biology-inspired joint distribution neurons based on Hierarchical
Correlation Reconstruction allowing for multidirectional neural networks | Biological neural networks seem qualitatively superior (e.g. in learning, flexibility, robustness) from current artificial like Multi-Layer Perceptron (MLP) or Kolmogorov-Arnold Network (KAN). Simultaneously, in contrast to them: have fundamentally multidirectional signal propagation~cite{axon}, also of probability distributions e.g. for uncertainty estimation, and are believed not being able to use standard backpropagation training~cite{backprop}. There are proposed novel artificial neurons based on HCR (Hierarchical Correlation Reconstruction) removing the above low level differences: with neurons containing local joint distribution model (of its connections), representing joint density on normalized variables as just linear combination among $(f_mathbf{j})$ orthonormal polynomials: $rho(mathbf{x})=sum_{mathbf{j}in B} a_mathbf{j} f_mathbf{j}(mathbf{x})$ for $mathbf{x} in [0,1]^d$ and $B$ some chosen basis, with basis growth approaching complete description of joint distribution. By various index summations of such $(a_mathbf{j})$ tensor as neuron parameters, we get simple formulas for e.g. conditional expected values for propagation in any direction, like $E[x|y,z]$, $E[y|x]$, which degenerate to KAN-like parametrization if restricting to pairwise dependencies. Such HCR network can also propagate probability distributions (also joint) like $rho(y,z|x)$. It also allows for additional training approaches, like direct $(a_mathbf{j})$ estimation, through tensor decomposition, or more biologically plausible information bottleneck training: layers directly influencing only neighbors, optimizing content to maximize information about the next layer, and minimizing about the previous to minimize the noise. | [
"['Jarek Duda']"
]
|
null | null | 2405.05134 | null | null | http://arxiv.org/pdf/2405.05134v1 | 2024-04-25T00:23:20Z | 2024-04-25T00:23:20Z | Enhancing Deep Knowledge Tracing via Diffusion Models for Personalized
Adaptive Learning | In contrast to pedagogies like evidence-based teaching, personalized adaptive learning (PAL) distinguishes itself by closely monitoring the progress of individual students and tailoring the learning path to their unique knowledge and requirements. A crucial technique for effective PAL implementation is knowledge tracing, which models students' evolving knowledge to predict their future performance. Based on these predictions, personalized recommendations for resources and learning paths can be made to meet individual needs. Recent advancements in deep learning have successfully enhanced knowledge tracking through Deep Knowledge Tracing (DKT). This paper introduces generative AI models to further enhance DKT. Generative AI models, rooted in deep learning, are trained to generate synthetic data, addressing data scarcity challenges in various applications across fields such as natural language processing (NLP) and computer vision (CV). This study aims to tackle data shortage issues in student learning records to enhance DKT performance for PAL. Specifically, it employs TabDDPM, a diffusion model, to generate synthetic educational records to augment training data for enhancing DKT. The proposed method's effectiveness is validated through extensive experiments on ASSISTments datasets. The experimental results demonstrate that the AI-generated data by TabDDPM significantly improves DKT performance, particularly in scenarios with small data for training and large data for testing. | [
"['Ming Kuo' 'Shouvon Sarker' 'Lijun Qian' 'Yujian Fu' 'Xiangfang Li'\n 'Xishuang Dong']"
]
|
null | null | 2405.05135 | null | null | http://arxiv.org/pdf/2405.05135v1 | 2024-04-24T20:26:48Z | 2024-04-24T20:26:48Z | Lessons from the Use of Natural Language Inference (NLI) in Requirements
Engineering Tasks | We investigate the use of Natural Language Inference (NLI) in automating requirements engineering tasks. In particular, we focus on three tasks: requirements classification, identification of requirements specification defects, and detection of conflicts in stakeholders' requirements. While previous research has demonstrated significant benefit in using NLI as a universal method for a broad spectrum of natural language processing tasks, these advantages have not been investigated within the context of software requirements engineering. Therefore, we design experiments to evaluate the use of NLI in requirements analysis. We compare the performance of NLI with a spectrum of approaches, including prompt-based models, conventional transfer learning, Large Language Models (LLMs)-powered chatbot models, and probabilistic models. Through experiments conducted under various learning settings including conventional learning and zero-shot, we demonstrate conclusively that our NLI method surpasses classical NLP methods as well as other LLMs-based and chatbot models in the analysis of requirements specifications. Additionally, we share lessons learned characterizing the learning settings that make NLI a suitable approach for automating requirements engineering tasks. | [
"['Mohamad Fazelnia' 'Viktoria Koscinski' 'Spencer Herzog'\n 'Mehdi Mirakhorli']"
]
|
null | null | 2405.05136 | null | null | http://arxiv.org/pdf/2405.05136v1 | 2024-04-24T18:19:44Z | 2024-04-24T18:19:44Z | Integrating LSTM and BERT for Long-Sequence Data Analysis in Intelligent
Tutoring Systems | The field of Knowledge Tracing aims to understand how students learn and master knowledge over time by analyzing their historical behaviour data. To achieve this goal, many researchers have proposed Knowledge Tracing models that use data from Intelligent Tutoring Systems to predict students' subsequent actions. However, with the development of Intelligent Tutoring Systems, large-scale datasets containing long-sequence data began to emerge. Recent deep learning based Knowledge Tracing models face obstacles such as low efficiency, low accuracy, and low interpretability when dealing with large-scale datasets containing long-sequence data. To address these issues and promote the sustainable development of Intelligent Tutoring Systems, we propose a LSTM BERT-based Knowledge Tracing model for long sequence data processing, namely LBKT, which uses a BERT-based architecture with a Rasch model-based embeddings block to deal with different difficulty levels information and an LSTM block to process the sequential characteristic in students' actions. LBKT achieves the best performance on most benchmark datasets on the metrics of ACC and AUC. Additionally, an ablation study is conducted to analyse the impact of each component of LBKT's overall performance. Moreover, we used t-SNE as the visualisation tool to demonstrate the model's embedding strategy. The results indicate that LBKT is faster, more interpretable, and has a lower memory cost than the traditional deep learning based Knowledge Tracing methods. | [
"['Zhaoxing Li' 'Jujie Yang' 'Jindi Wang' 'Lei Shi' 'Sebastian Stein']"
]
|
null | null | 2405.05140 | null | null | http://arxiv.org/pdf/2405.05140v1 | 2024-04-22T16:37:35Z | 2024-04-22T16:37:35Z | Distributed Learning for Wi-Fi AP Load Prediction | The increasing cloudification and softwarization of networks foster the interplay among multiple independently managed deployments. An appealing reason for such an interplay lies in distributed Machine Learning (ML), which allows the creation of robust ML models by leveraging collective intelligence and computational power. In this paper, we study the application of the two cornerstones of distributed learning, namely Federated Learning (FL) and Knowledge Distillation (KD), on the Wi-Fi Access Point (AP) load prediction use case. The analysis conducted in this paper is done on a dataset that contains real measurements from a large Wi-Fi campus network, which we use to train the ML model under study based on different strategies. Performance evaluation includes relevant aspects for the suitability of distributed learning operation in real use cases, including the predictive performance, the associated communication overheads, or the energy consumption. In particular, we prove that distributed learning can improve the predictive accuracy centralized ML solutions by up to 93% while reducing the communication overheads and the energy cost by 80%. | [
"['Dariush Salami' 'Francesc Wilhelmi' 'Lorenzo Galati-Giordano'\n 'Mika Kasslin']"
]
|
null | null | 2405.05141 | null | null | http://arxiv.org/pdf/2405.05141v1 | 2024-04-22T15:03:46Z | 2024-04-22T15:03:46Z | Learning-to-learn enables rapid learning with phase-change memory-based
in-memory computing | There is a growing demand for low-power, autonomously learning artificial intelligence (AI) systems that can be applied at the edge and rapidly adapt to the specific situation at deployment site. However, current AI models struggle in such scenarios, often requiring extensive fine-tuning, computational resources, and data. In contrast, humans can effortlessly adjust to new tasks by transferring knowledge from related ones. The concept of learning-to-learn (L2L) mimics this process and enables AI models to rapidly adapt with only little computational effort and data. In-memory computing neuromorphic hardware (NMHW) is inspired by the brain's operating principles and mimics its physical co-location of memory and compute. In this work, we pair L2L with in-memory computing NMHW based on phase-change memory devices to build efficient AI models that can rapidly adapt to new tasks. We demonstrate the versatility of our approach in two scenarios: a convolutional neural network performing image classification and a biologically-inspired spiking neural network generating motor commands for a real robotic arm. Both models rapidly learn with few parameter updates. Deployed on the NMHW, they perform on-par with their software equivalents. Moreover, meta-training of these models can be performed in software with high-precision, alleviating the need for accurate hardware models. | [
"['Thomas Ortner' 'Horst Petschenig' 'Athanasios Vasilopoulos'\n 'Roland Renner' 'Špela Brglez' 'Thomas Limbacher' 'Enrique Piñero'\n 'Alejandro Linares Barranco' 'Angeliki Pantazi' 'Robert Legenstein']"
]
|
null | null | 2405.05142 | null | null | http://arxiv.org/pdf/2405.05142v1 | 2024-04-20T02:34:03Z | 2024-04-20T02:34:03Z | Ordinal Behavior Classification of Student Online Course Interactions | The study in interaction patterns between students in on-campus and MOOC-style online courses has been broadly studied for the last 11 years. Yet there remains a gap in the literature comparing the habits of students completing the same course offered in both on-campus and MOOC-style online formats. This study will look at browser-based usage patterns for students in the Georgia Tech CS1301 edx course for both the online course offered to on-campus students and the MOOCstyle course offered to anyone to determine what, if any, patterns exist between the two cohorts. | [
"['Thomas Trask']"
]
|
null | null | 2405.05143 | null | null | http://arxiv.org/pdf/2405.05143v1 | 2024-04-19T14:08:17Z | 2024-04-19T14:08:17Z | Learning Object Semantic Similarity with Self-Supervision | Humans judge the similarity of two objects not just based on their visual appearance but also based on their semantic relatedness. However, it remains unclear how humans learn about semantic relationships between objects and categories. One important source of semantic knowledge is that semantically related objects frequently co-occur in the same context. For instance, forks and plates are perceived as similar, at least in part, because they are often experienced together in a ``kitchen" or ``eating'' context. Here, we investigate whether a bio-inspired learning principle exploiting such co-occurrence statistics suffices to learn a semantically structured object representation {em de novo} from raw visual or combined visual and linguistic input. To this end, we simulate temporal sequences of visual experience by binding together short video clips of real-world scenes showing objects in different contexts. A bio-inspired neural network model aligns close-in-time visual representations while also aligning visual and category label representations to simulate visuo-language alignment. Our results show that our model clusters object representations based on their context, e.g. kitchen or bedroom, in particular in high-level layers of the network, akin to humans. In contrast, lower-level layers tend to better reflect object identity or category. To achieve this, the model exploits two distinct strategies: the visuo-language alignment ensures that different objects of the same category are represented similarly, whereas the temporal alignment leverages that objects from the same context are frequently seen in succession to make their representations more similar. Overall, our work suggests temporal and visuo-language alignment as plausible computational principles for explaining the origins of certain forms of semantic knowledge in humans. | [
"['Arthur Aubret' 'Timothy Schaumlöffel' 'Gemma Roig' 'Jochen Triesch']"
]
|
null | null | 2405.05144 | null | null | http://arxiv.org/pdf/2405.05144v2 | 2024-05-13T18:10:19Z | 2024-04-19T00:25:44Z | Improving Automated Distractor Generation for Math Multiple-choice
Questions with Overgenerate-and-rank | Multiple-choice questions (MCQs) are commonly used across all levels of math education since they can be deployed and graded at a large scale. A critical component of MCQs is the distractors, i.e., incorrect answers crafted to reflect student errors or misconceptions. Automatically generating them in math MCQs, e.g., with large language models, has been challenging. In this work, we propose a novel method to enhance the quality of generated distractors through overgenerate-and-rank, training a ranking model to predict how likely distractors are to be selected by real students. Experimental results on a real-world dataset and human evaluation with math teachers show that our ranking model increases alignment with human-authored distractors, although human-authored ones are still preferred over generated ones. | [
"['Alexander Scarlatos' 'Wanyong Feng' 'Digory Smith' 'Simon Woodhead'\n 'Andrew Lan']"
]
|
null | null | 2405.05145 | null | null | http://arxiv.org/pdf/2405.05145v1 | 2024-04-16T15:51:39Z | 2024-04-16T15:51:39Z | Conformal Semantic Image Segmentation: Post-hoc Quantification of
Predictive Uncertainty | We propose a post-hoc, computationally lightweight method to quantify predictive uncertainty in semantic image segmentation. Our approach uses conformal prediction to generate statistically valid prediction sets that are guaranteed to include the ground-truth segmentation mask at a predefined confidence level. We introduce a novel visualization technique of conformalized predictions based on heatmaps, and provide metrics to assess their empirical validity. We demonstrate the effectiveness of our approach on well-known benchmark datasets and image segmentation prediction models, and conclude with practical insights. | [
"['Luca Mossina' 'Joseba Dalmau' 'Léo andéol']"
]
|
null | null | 2405.05160 | null | null | http://arxiv.org/pdf/2405.05160v1 | 2024-05-08T15:52:50Z | 2024-05-08T15:52:50Z | Selective Classification Under Distribution Shifts | In selective classification (SC), a classifier abstains from making predictions that are likely to be wrong to avoid excessive errors. To deploy imperfect classifiers -- imperfect either due to intrinsic statistical noise of data or for robustness issue of the classifier or beyond -- in high-stakes scenarios, SC appears to be an attractive and necessary path to follow. Despite decades of research in SC, most previous SC methods still focus on the ideal statistical setting only, i.e., the data distribution at deployment is the same as that of training, although practical data can come from the wild. To bridge this gap, in this paper, we propose an SC framework that takes into account distribution shifts, termed generalized selective classification, that covers label-shifted (or out-of-distribution) and covariate-shifted samples, in addition to typical in-distribution samples, the first of its kind in the SC literature. We focus on non-training-based confidence-score functions for generalized SC on deep learning (DL) classifiers and propose two novel margin-based score functions. Through extensive analysis and experiments, we show that our proposed score functions are more effective and reliable than the existing ones for generalized SC on a variety of classification tasks and DL classifiers. | [
"['Hengyue Liang' 'Le Peng' 'Ju Sun']"
]
|
null | null | 2405.05167 | null | null | http://arxiv.org/pdf/2405.05167v1 | 2024-05-08T16:04:50Z | 2024-05-08T16:04:50Z | Data-Error Scaling in Machine Learning on Natural Discrete Combinatorial
Mutation-prone Sets: Case Studies on Peptides and Small Molecules | We investigate trends in the data-error scaling behavior of machine learning (ML) models trained on discrete combinatorial spaces that are prone-to-mutation, such as proteins or organic small molecules. We trained and evaluated kernel ridge regression machines using variable amounts of computationally generated training data. Our synthetic datasets comprise i) two na"ive functions based on many-body theory; ii) binding energy estimates between a protein and a mutagenised peptide; and iii) solvation energies of two 6-heavy atom structural graphs. In contrast to typical data-error scaling, our results showed discontinuous monotonic phase transitions during learning, observed as rapid drops in the test error at particular thresholds of training data. We observed two learning regimes, which we call saturated and asymptotic decay, and found that they are conditioned by the level of complexity (i.e. number of mutations) enclosed in the training set. We show that during training on this class of problems, the predictions were clustered by the ML models employed in the calibration plots. Furthermore, we present an alternative strategy to normalize learning curves (LCs) and the concept of mutant based shuffling. This work has implications for machine learning on mutagenisable discrete spaces such as chemical properties or protein phenotype prediction, and improves basic understanding of concepts in statistical learning theory. | [
"['Vanni Doffini' 'O. Anatole von Lilienfeld' 'Michael A. Nash']"
]
|
null | null | 2405.05171 | null | null | http://arxiv.org/pdf/2405.05171v3 | 2024-05-22T19:36:06Z | 2024-05-08T16:07:56Z | Custom Gradient Estimators are Straight-Through Estimators in Disguise | Quantization-aware training comes with a fundamental challenge: the derivative of quantization functions such as rounding are zero almost everywhere and nonexistent elsewhere. Various differentiable approximations of quantization functions have been proposed to address this issue. In this paper, we prove that when the learning rate is sufficiently small, a large class of weight gradient estimators is equivalent with the straight through estimator (STE). Specifically, after swapping in the STE and adjusting both the weight initialization and the learning rate in SGD, the model will train in almost exactly the same way as it did with the original gradient estimator. Moreover, we show that for adaptive learning rate algorithms like Adam, the same result can be seen without any modifications to the weight initialization and learning rate. We experimentally show that these results hold for both a small convolutional model trained on the MNIST dataset and for a ResNet50 model trained on ImageNet. | [
"['Matt Schoenbauer' 'Daniele Moro' 'Lukasz Lew' 'Andrew Howard']"
]
|
null | null | 2405.05175 | null | null | http://arxiv.org/pdf/2405.05175v1 | 2024-05-08T16:12:45Z | 2024-05-08T16:12:45Z | Air Gap: Protecting Privacy-Conscious Conversational Agents | The growing use of large language model (LLM)-based conversational agents to manage sensitive user data raises significant privacy concerns. While these agents excel at understanding and acting on context, this capability can be exploited by malicious actors. We introduce a novel threat model where adversarial third-party apps manipulate the context of interaction to trick LLM-based agents into revealing private information not relevant to the task at hand. Grounded in the framework of contextual integrity, we introduce AirGapAgent, a privacy-conscious agent designed to prevent unintended data leakage by restricting the agent's access to only the data necessary for a specific task. Extensive experiments using Gemini, GPT, and Mistral models as agents validate our approach's effectiveness in mitigating this form of context hijacking while maintaining core agent functionality. For example, we show that a single-query context hijacking attack on a Gemini Ultra agent reduces its ability to protect user data from 94% to 45%, while an AirGapAgent achieves 97% protection, rendering the same attack ineffective. | [
"['Eugene Bagdasaryan' 'Ren Yi' 'Sahra Ghalebikesabi' 'Peter Kairouz'\n 'Marco Gruteser' 'Sewoong Oh' 'Borja Balle' 'Daniel Ramage']"
]
|
null | null | 2405.05185 | null | null | http://arxiv.org/pdf/2405.05185v1 | 2024-05-08T16:20:47Z | 2024-05-08T16:20:47Z | Machine Learning Assisted Dynamical Classification of Trans-Neptunian
Objects | Trans-Neptunian objects (TNOs) are small, icy bodies in the outer solar system. They are observed to have a complex orbital distribution that was shaped by the early dynamical history and migration of the giant planets. Comparisons between the different dynamical classes of modeled and observed TNOs can help constrain the history of the outer solar system. Because of the complex dynamics of TNOs, particularly those in and near mean motion resonances with Neptune, classification has traditionally been done by human inspection of plots of the time evolution of orbital parameters. This is very inefficient. The Vera Rubin Observatory's Legacy Survey of Space and Time (LSST) is expected to increase the number of known TNOs by a factor of $sim$10, necessitating a much more automated process. In this chapter we present an improved supervised machine learning classifier for TNOs. Using a large and diverse training set as well as carefully chosen, dynamically motivated data features calculated from numerical integrations of TNO orbits, our classifier returns results that match those of a human classifier 98% of the time, and dynamically relevant classifications 99.7% of the time. This classifier is dramatically more efficient than human classification, and it will improve classification of both observed and modeled TNO data. | [
"['Kathryn Volk' 'Renu Malhotra']"
]
|
null | null | 2405.05187 | null | null | http://arxiv.org/pdf/2405.05187v1 | 2024-05-08T16:22:47Z | 2024-05-08T16:22:47Z | A score-based particle method for homogeneous Landau equation | We propose a novel score-based particle method for solving the Landau equation in plasmas, that seamlessly integrates learning with structure-preserving particle methods [arXiv:1910.03080]. Building upon the Lagrangian viewpoint of the Landau equation, a central challenge stems from the nonlinear dependence of the velocity field on the density. Our primary innovation lies in recognizing that this nonlinearity is in the form of the score function, which can be approximated dynamically via techniques from score-matching. The resulting method inherits the conservation properties of the deterministic particle method while sidestepping the necessity for kernel density estimation in [arXiv:1910.03080]. This streamlines computation and enhances scalability with dimensionality. Furthermore, we provide a theoretical estimate by demonstrating that the KL divergence between our approximation and the true solution can be effectively controlled by the score-matching loss. Additionally, by adopting the flow map viewpoint, we derive an update formula for exact density computation. Extensive examples have been provided to show the efficiency of the method, including a physically relevant case of Coulomb interaction. | [
"['Yan Huang' 'Li Wang']"
]
|
null | null | 2405.05190 | null | null | http://arxiv.org/pdf/2405.05190v1 | 2024-05-08T16:26:49Z | 2024-05-08T16:26:49Z | Is Transductive Learning Equivalent to PAC Learning? | Most work in the area of learning theory has focused on designing effective Probably Approximately Correct (PAC) learners. Recently, other models of learning such as transductive error have seen more scrutiny. We move toward showing that these problems are equivalent by reducing agnostic learning with a PAC guarantee to agnostic learning with a transductive guarantee by adding a small number of samples to the dataset. We first rederive the result of Aden-Ali et al. arXiv:2304.09167 reducing PAC learning to transductive learning in the realizable setting using simpler techniques and at more generality as background for our main positive result. Our agnostic transductive to PAC conversion technique extends the aforementioned argument to the agnostic case, showing that an agnostic transductive learner can be efficiently converted to an agnostic PAC learner. Finally, we characterize the performance of the agnostic one inclusion graph algorithm of Asilis et al. arXiv:2309.13692 for binary classification, and show that plugging it into our reduction leads to an agnostic PAC learner that is essentially optimal. Our results imply that transductive and PAC learning are essentially equivalent for supervised learning with pseudometric losses in the realizable setting, and for binary classification in the agnostic setting. We conjecture this is true more generally for the agnostic setting. | [
"['Shaddin Dughmi' 'Yusuf Kalayci' 'Grayson York']"
]
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.