categories
string | doi
string | id
string | year
float64 | venue
string | link
string | updated
string | published
string | title
string | abstract
string | authors
list |
|---|---|---|---|---|---|---|---|---|---|---|
null | null | 2404.08376 | null | null | http://arxiv.org/pdf/2404.08376v1 | 2024-04-12T10:22:55Z | 2024-04-12T10:22:55Z | Graph data augmentation with Gromow-Wasserstein Barycenters | Graphs are ubiquitous in various fields, and deep learning methods have been successful applied in graph classification tasks. However, building large and diverse graph datasets for training can be expensive. While augmentation techniques exist for structured data like images or numerical data, the augmentation of graph data remains challenging. This is primarily due to the complex and non-Euclidean nature of graph data. In this paper, it has been proposed a novel augmentation strategy for graphs that operates in a non-Euclidean space. This approach leverages graphon estimation, which models the generative mechanism of networks sequences. Computational results demonstrate the effectiveness of the proposed augmentation framework in improving the performance of graph classification models. Additionally, using a non-Euclidean distance, specifically the Gromow-Wasserstein distance, results in better approximations of the graphon. This framework also provides a means to validate different graphon estimation approaches, particularly in real-world scenarios where the true graphon is unknown. | [
"['Andrea Ponti']"
]
|
null | null | 2404.08392 | null | null | http://arxiv.org/pdf/2404.08392v1 | 2024-04-12T10:54:11Z | 2024-04-12T10:54:11Z | NC-TTT: A Noise Contrastive Approach for Test-Time Training | Despite their exceptional performance in vision tasks, deep learning models often struggle when faced with domain shifts during testing. Test-Time Training (TTT) methods have recently gained popularity by their ability to enhance the robustness of models through the addition of an auxiliary objective that is jointly optimized with the main task. Being strictly unsupervised, this auxiliary objective is used at test time to adapt the model without any access to labels. In this work, we propose Noise-Contrastive Test-Time Training (NC-TTT), a novel unsupervised TTT technique based on the discrimination of noisy feature maps. By learning to classify noisy views of projected feature maps, and then adapting the model accordingly on new domains, classification performance can be recovered by an important margin. Experiments on several popular test-time adaptation baselines demonstrate the advantages of our method compared to recent approaches for this task. The code can be found at:https://github.com/GustavoVargasHakim/NCTTT.git | [
"['David Osowiechi' 'Gustavo A. Vargas Hakim' 'Mehrdad Noori'\n 'Milad Cheraghalikhani' 'Ali Bahri' 'Moslem Yazdanpanah'\n 'Ismail Ben Ayed' 'Christian Desrosiers']"
]
|
null | null | 2404.08397 | null | null | http://arxiv.org/pdf/2404.08397v1 | 2024-04-12T11:06:22Z | 2024-04-12T11:06:22Z | Data-Driven Preference Sampling for Pareto Front Learning | Pareto front learning is a technique that introduces preference vectors in a neural network to approximate the Pareto front. Previous Pareto front learning methods have demonstrated high performance in approximating simple Pareto fronts. These methods often sample preference vectors from a fixed Dirichlet distribution. However, no fixed sampling distribution can be adapted to diverse Pareto fronts. Efficiently sampling preference vectors and accurately estimating the Pareto front is a challenge. To address this challenge, we propose a data-driven preference vector sampling framework for Pareto front learning. We utilize the posterior information of the objective functions to adjust the parameters of the sampling distribution flexibly. In this manner, the proposed method can sample preference vectors from the location of the Pareto front with a high probability. Moreover, we design the distribution of the preference vector as a mixture of Dirichlet distributions to improve the performance of the model in disconnected Pareto fronts. Extensive experiments validate the superiority of the proposed method compared with state-of-the-art algorithms. | [
"['Rongguang Ye' 'Lei Chen' 'Weiduo Liao' 'Jinyuan Zhang' 'Hisao Ishibuchi']"
]
|
null | null | 2404.08403 | null | null | http://arxiv.org/pdf/2404.08403v1 | 2024-04-12T11:30:16Z | 2024-04-12T11:30:16Z | Learning representations of learning representations | The ICLR conference is unique among the top machine learning conferences in that all submitted papers are openly available. Here we present the ICLR dataset consisting of abstracts of all 24 thousand ICLR submissions from 2017-2024 with meta-data, decision scores, and custom keyword-based labels. We find that on this dataset, bag-of-words representation outperforms most dedicated sentence transformer models in terms of $k$NN classification accuracy, and the top performing language models barely outperform TF-IDF. We see this as a challenge for the NLP community. Furthermore, we use the ICLR dataset to study how the field of machine learning has changed over the last seven years, finding some improvement in gender balance. Using a 2D embedding of the abstracts' texts, we describe a shift in research topics from 2017 to 2024 and identify hedgehogs and foxes among the authors with the highest number of ICLR submissions. | [
"['Rita González-Márquez' 'Dmitry Kobak']"
]
|
null | null | 2404.08404 | null | null | http://arxiv.org/pdf/2404.08404v1 | 2024-04-12T11:31:37Z | 2024-04-12T11:31:37Z | Complexity of Probabilistic Reasoning for Neurosymbolic Classification
Techniques | Neurosymbolic artificial intelligence is a growing field of research aiming to combine neural network learning capabilities with the reasoning abilities of symbolic systems. Informed multi-label classification is a sub-field of neurosymbolic AI which studies how to leverage prior knowledge to improve neural classification systems. A well known family of neurosymbolic techniques for informed classification use probabilistic reasoning to integrate this knowledge during learning, inference or both. Therefore, the asymptotic complexity of probabilistic reasoning is of cardinal importance to assess the scalability of such techniques. However, this topic is rarely tackled in the neurosymbolic literature, which can lead to a poor understanding of the limits of probabilistic neurosymbolic techniques. In this paper, we introduce a formalism for informed supervised classification tasks and techniques. We then build upon this formalism to define three abstract neurosymbolic techniques based on probabilistic reasoning. Finally, we show computational complexity results on several representation languages for prior knowledge commonly found in the neurosymbolic literature. | [
"['Arthur Ledaguenel' 'Céline Hudelot' 'Mostepha Khouadjia']"
]
|
null | null | 2404.08408 | null | null | http://arxiv.org/pdf/2404.08408v1 | 2024-04-12T11:36:24Z | 2024-04-12T11:36:24Z | Seismic First Break Picking in a Higher Dimension Using Deep Graph
Learning | Contemporary automatic first break (FB) picking methods typically analyze 1D signals, 2D source gathers, or 3D source-receiver gathers. Utilizing higher-dimensional data, such as 2D or 3D, incorporates global features, improving the stability of local picking. Despite the benefits, high-dimensional data requires structured input and increases computational demands. Addressing this, we propose a novel approach using deep graph learning called DGL-FB, constructing a large graph to efficiently extract information. In this graph, each seismic trace is represented as a node, connected by edges that reflect similarities. To manage the size of the graph, we develop a subgraph sampling technique to streamline model training and inference. Our proposed framework, DGL-FB, leverages deep graph learning for FB picking. It encodes subgraphs into global features using a deep graph encoder. Subsequently, the encoded global features are combined with local node signals and fed into a ResUNet-based 1D segmentation network for FB detection. Field survey evaluations of DGL-FB show superior accuracy and stability compared to a 2D U-Net-based benchmark method. | [
"['Hongtao Wang' 'Li Long' 'Jiangshe Zhang' 'Xiaoli Wei' 'Chunxia Zhang'\n 'Zhenbo Guo']"
]
|
null | null | 2404.08417 | null | null | http://arxiv.org/pdf/2404.08417v1 | 2024-04-12T12:06:02Z | 2024-04-12T12:06:02Z | AdapterSwap: Continuous Training of LLMs with Data Removal and
Access-Control Guarantees | Large language models (LLMs) are increasingly capable of completing knowledge intensive tasks by recalling information from a static pretraining corpus. Here we are concerned with LLMs in the context of evolving data requirements. For instance: batches of new data that are introduced periodically; subsets of data with user-based access controls; or requirements on dynamic removal of documents with guarantees that associated knowledge cannot be recalled. We wish to satisfy these requirements while at the same time ensuring a model does not forget old information when new data becomes available. To address these issues, we introduce AdapterSwap, a training and inference scheme that organizes knowledge from a data collection into a set of low-rank adapters, which are dynamically composed during inference. Our experiments demonstrate AdapterSwap's ability to support efficient continual learning, while also enabling organizations to have fine-grained control over data access and deletion. | [
"['William Fleshman' 'Aleem Khan' 'Marc Marone' 'Benjamin Van Durme']"
]
|
null | null | 2404.08423 | null | null | http://arxiv.org/pdf/2404.08423v2 | 2024-04-30T12:28:21Z | 2024-04-12T12:11:51Z | SIR-RL: Reinforcement Learning for Optimized Policy Control during
Epidemiological Outbreaks in Emerging Market and Developing Economies | The outbreak of COVID-19 has highlighted the intricate interplay between public health and economic stability on a global scale. This study proposes a novel reinforcement learning framework designed to optimize health and economic outcomes during pandemics. The framework leverages the SIR model, integrating both lockdown measures (via a stringency index) and vaccination strategies to simulate disease dynamics. The stringency index, indicative of the severity of lockdown measures, influences both the spread of the disease and the economic health of a country. Developing nations, which bear a disproportionate economic burden under stringent lockdowns, are the primary focus of our study. By implementing reinforcement learning, we aim to optimize governmental responses and strike a balance between the competing costs associated with public health and economic stability. This approach also enhances transparency in governmental decision-making by establishing a well-defined reward function for the reinforcement learning agent. In essence, this study introduces an innovative and ethical strategy to navigate the challenge of balancing public health and economic stability amidst infectious disease outbreaks. | [
"['Maeghal Jain' 'Ziya Uddin' 'Wubshet Ibrahim']"
]
|
null | null | 2404.08434 | null | null | http://arxiv.org/pdf/2404.08434v1 | 2024-04-12T12:31:06Z | 2024-04-12T12:31:06Z | An improved tabular data generator with VAE-GMM integration | The rising use of machine learning in various fields requires robust methods to create synthetic tabular data. Data should preserve key characteristics while addressing data scarcity challenges. Current approaches based on Generative Adversarial Networks, such as the state-of-the-art CTGAN model, struggle with the complex structures inherent in tabular data. These data often contain both continuous and discrete features with non-Gaussian distributions. Therefore, we propose a novel Variational Autoencoder (VAE)-based model that addresses these limitations. Inspired by the TVAE model, our approach incorporates a Bayesian Gaussian Mixture model (BGM) within the VAE architecture. This avoids the limitations imposed by assuming a strictly Gaussian latent space, allowing for a more accurate representation of the underlying data distribution during data generation. Furthermore, our model offers enhanced flexibility by allowing the use of various differentiable distributions for individual features, making it possible to handle both continuous and discrete data types. We thoroughly validate our model on three real-world datasets with mixed data types, including two medically relevant ones, based on their resemblance and utility. This evaluation demonstrates significant outperformance against CTGAN and TVAE, establishing its potential as a valuable tool for generating synthetic tabular data in various domains, particularly in healthcare. | [
"['Patricia A. Apellániz' 'Juan Parras' 'Santiago Zazo']"
]
|
null | null | 2404.08444 | null | null | http://arxiv.org/pdf/2404.08444v1 | 2024-04-12T12:56:16Z | 2024-04-12T12:56:16Z | Anti-Byzantine Attacks Enabled Vehicle Selection for Asynchronous
Federated Learning in Vehicular Edge Computing | In vehicle edge computing (VEC), asynchronous federated learning (AFL) is used, where the edge receives a local model and updates the global model, effectively reducing the global aggregation latency.Due to different amounts of local data,computing capabilities and locations of the vehicles, renewing the global model with same weight is inappropriate.The above factors will affect the local calculation time and upload time of the local model, and the vehicle may also be affected by Byzantine attacks, leading to the deterioration of the vehicle data. However, based on deep reinforcement learning (DRL), we can consider these factors comprehensively to eliminate vehicles with poor performance as much as possible and exclude vehicles that have suffered Byzantine attacks before AFL. At the same time, when aggregating AFL, we can focus on those vehicles with better performance to improve the accuracy and safety of the system. In this paper, we proposed a vehicle selection scheme based on DRL in VEC. In this scheme, vehicle s mobility, channel conditions with temporal variations, computational resources with temporal variations, different data amount, transmission channel status of vehicles as well as Byzantine attacks were taken into account.Simulation results show that the proposed scheme effectively improves the safety and accuracy of the global model. | [
"['Cui Zhang' 'Xiao Xu' 'Qiong Wu' 'Pingyi Fan' 'Qiang Fan' 'Huiling Zhu'\n 'Jiangzhou Wang']"
]
|
null | null | 2404.08447 | null | null | http://arxiv.org/pdf/2404.08447v1 | 2024-04-12T12:57:43Z | 2024-04-12T12:57:43Z | Federated Optimization with Doubly Regularized Drift Correction | Federated learning is a distributed optimization paradigm that allows training machine learning models across decentralized devices while keeping the data localized. The standard method, FedAvg, suffers from client drift which can hamper performance and increase communication costs over centralized methods. Previous works proposed various strategies to mitigate drift, yet none have shown uniformly improved communication-computation trade-offs over vanilla gradient descent. In this work, we revisit DANE, an established method in distributed optimization. We show that (i) DANE can achieve the desired communication reduction under Hessian similarity constraints. Furthermore, (ii) we present an extension, DANE+, which supports arbitrary inexact local solvers and has more freedom to choose how to aggregate the local updates. We propose (iii) a novel method, FedRed, which has improved local computational complexity and retains the same communication complexity compared to DANE/DANE+. This is achieved by using doubly regularized drift correction. | [
"['Xiaowen Jiang' 'Anton Rodomanov' 'Sebastian U. Stich']"
]
|
null | null | 2404.08453 | null | null | http://arxiv.org/pdf/2404.08453v1 | 2024-04-12T13:02:33Z | 2024-04-12T13:02:33Z | Lightweight Multi-System Multivariate Interconnection and Divergence
Discovery | Identifying outlier behavior among sensors and subsystems is essential for discovering faults and facilitating diagnostics in large systems. At the same time, exploring large systems with numerous multivariate data sets is challenging. This study presents a lightweight interconnection and divergence discovery mechanism (LIDD) to identify abnormal behavior in multi-system environments. The approach employs a multivariate analysis technique that first estimates the similarity heatmaps among the sensors for each system and then applies information retrieval algorithms to provide relevant multi-level interconnection and discrepancy details. Our experiment on the readout systems of the Hadron Calorimeter of the Compact Muon Solenoid (CMS) experiment at CERN demonstrates the effectiveness of the proposed method. Our approach clusters readout systems and their sensors consistent with the expected calorimeter interconnection configurations, while capturing unusual behavior in divergent clusters and estimating their root causes. | [
"['Mulugeta Weldezgina Asres' 'Christian Walter Omlin' 'Jay Dittmann'\n 'Pavel Parygin' 'Joshua Hiltbrand' 'Seth I. Cooper' 'Grace Cummings'\n 'David Yu']"
]
|
null | null | 2404.08456 | null | null | http://arxiv.org/pdf/2404.08456v1 | 2024-04-12T13:05:35Z | 2024-04-12T13:05:35Z | A backward differential deep learning-based algorithm for solving
high-dimensional nonlinear backward stochastic differential equations | In this work, we propose a novel backward differential deep learning-based algorithm for solving high-dimensional nonlinear backward stochastic differential equations (BSDEs), where the deep neural network (DNN) models are trained not only on the inputs and labels but also the differentials of the corresponding labels. This is motivated by the fact that differential deep learning can provide an efficient approximation of the labels and their derivatives with respect to inputs. The BSDEs are reformulated as differential deep learning problems by using Malliavin calculus. The Malliavin derivatives of solution to a BSDE satisfy themselves another BSDE, resulting thus in a system of BSDEs. Such formulation requires the estimation of the solution, its gradient, and the Hessian matrix, represented by the triple of processes $left(Y, Z, Gammaright).$ All the integrals within this system are discretized by using the Euler-Maruyama method. Subsequently, DNNs are employed to approximate the triple of these unknown processes. The DNN parameters are backwardly optimized at each time step by minimizing a differential learning type loss function, which is defined as a weighted sum of the dynamics of the discretized BSDE system, with the first term providing the dynamics of the process $Y$ and the other the process $Z$. An error analysis is carried out to show the convergence of the proposed algorithm. Various numerical experiments up to $50$ dimensions are provided to demonstrate the high efficiency. Both theoretically and numerically, it is demonstrated that our proposed scheme is more efficient compared to other contemporary deep learning-based methodologies, especially in the computation of the process $Gamma$. | [
"['Lorenc Kapllani' 'Long Teng']"
]
|
null | null | 2404.08458 | null | null | http://arxiv.org/pdf/2404.08458v2 | 2024-06-07T15:10:50Z | 2024-04-12T13:09:48Z | On the Independence Assumption in Neurosymbolic Learning | State-of-the-art neurosymbolic learning systems use probabilistic reasoning to guide neural networks towards predictions that conform to logical constraints over symbols. Many such systems assume that the probabilities of the considered symbols are conditionally independent given the input to simplify learning and reasoning. We study and criticise this assumption, highlighting how it can hinder optimisation and prevent uncertainty quantification. We prove that loss functions bias conditionally independent neural networks to become overconfident in their predictions. As a result, they are unable to represent uncertainty over multiple valid options. Furthermore, we prove that these loss functions are difficult to optimise: they are non-convex, and their minima are usually highly disconnected. Our theoretical analysis gives the foundation for replacing the conditional independence assumption and designing more expressive neurosymbolic probabilistic models. | [
"['Emile van Krieken' 'Pasquale Minervini' 'Edoardo M. Ponti'\n 'Antonio Vergari']"
]
|
null | null | 2404.08461 | null | null | http://arxiv.org/pdf/2404.08461v1 | 2024-04-12T13:18:47Z | 2024-04-12T13:18:47Z | OTTER: Improving Zero-Shot Classification via Optimal Transport | Popular zero-shot models suffer due to artifacts inherited from pretraining. A particularly detrimental artifact, caused by unbalanced web-scale pretraining data, is mismatched label distribution. Existing approaches that seek to repair the label distribution are not suitable in zero-shot settings, as they have incompatible requirements such as access to labeled downstream task data or knowledge of the true label balance in the pretraining distribution. We sidestep these challenges and introduce a simple and lightweight approach to adjust pretrained model predictions via optimal transport. Our technique requires only an estimate of the label distribution of a downstream task. Theoretically, we characterize the improvement produced by our procedure under certain mild conditions and provide bounds on the error caused by misspecification. Empirically, we validate our method in a wide array of zero-shot image and text classification tasks, improving accuracy by 4.8% and 15.9% on average, and beating baselines like Prior Matching -- often by significant margins -- in 17 out of 21 datasets. | [
"['Changho Shin' 'Jitian Zhao' 'Sonia Cromp' 'Harit Vishwakarma'\n 'Frederic Sala']"
]
|
null | null | 2404.08471 | null | null | http://arxiv.org/pdf/2404.08471v1 | 2024-02-15T18:59:11Z | 2024-02-15T18:59:11Z | Revisiting Feature Prediction for Learning Visual Representations from
Video | This paper explores feature prediction as a stand-alone objective for unsupervised learning from video and introduces V-JEPA, a collection of vision models trained solely using a feature prediction objective, without the use of pretrained image encoders, text, negative examples, reconstruction, or other sources of supervision. The models are trained on 2 million videos collected from public datasets and are evaluated on downstream image and video tasks. Our results show that learning by predicting video features leads to versatile visual representations that perform well on both motion and appearance-based tasks, without adaption of the model's parameters; e.g., using a frozen backbone. Our largest model, a ViT-H/16 trained only on videos, obtains 81.9% on Kinetics-400, 72.2% on Something-Something-v2, and 77.9% on ImageNet1K. | [
"['Adrien Bardes' 'Quentin Garrido' 'Jean Ponce' 'Xinlei Chen'\n 'Michael Rabbat' 'Yann LeCun' 'Mahmoud Assran' 'Nicolas Ballas']"
]
|
null | null | 2404.08472 | null | null | http://arxiv.org/pdf/2404.08472v2 | 2024-05-06T04:00:17Z | 2024-04-12T13:41:29Z | TSLANet: Rethinking Transformers for Time Series Representation Learning | Time series data, characterized by its intrinsic long and short-range dependencies, poses a unique challenge across analytical applications. While Transformer-based models excel at capturing long-range dependencies, they face limitations in noise sensitivity, computational efficiency, and overfitting with smaller datasets. In response, we introduce a novel Time Series Lightweight Adaptive Network (TSLANet), as a universal convolutional model for diverse time series tasks. Specifically, we propose an Adaptive Spectral Block, harnessing Fourier analysis to enhance feature representation and to capture both long-term and short-term interactions while mitigating noise via adaptive thresholding. Additionally, we introduce an Interactive Convolution Block and leverage self-supervised learning to refine the capacity of TSLANet for decoding complex temporal patterns and improve its robustness on different datasets. Our comprehensive experiments demonstrate that TSLANet outperforms state-of-the-art models in various tasks spanning classification, forecasting, and anomaly detection, showcasing its resilience and adaptability across a spectrum of noise levels and data sizes. The code is available at https://github.com/emadeldeen24/TSLANet. | [
"['Emadeldeen Eldele' 'Mohamed Ragab' 'Zhenghua Chen' 'Min Wu' 'Xiaoli Li']"
]
|
null | null | 2404.08476 | null | null | http://arxiv.org/pdf/2404.08476v1 | 2024-04-12T13:54:21Z | 2024-04-12T13:54:21Z | Combining Statistical Depth and Fermat Distance for Uncertainty
Quantification | We measure the Out-of-domain uncertainty in the prediction of Neural Networks using a statistical notion called ``Lens Depth'' (LD) combined with Fermat Distance, which is able to capture precisely the ``depth'' of a point with respect to a distribution in feature space, without any assumption about the form of distribution. Our method has no trainable parameter. The method is applicable to any classification model as it is applied directly in feature space at test time and does not intervene in training process. As such, it does not impact the performance of the original model. The proposed method gives excellent qualitative result on toy datasets and can give competitive or better uncertainty estimation on standard deep learning datasets compared to strong baseline methods. | [
"['Hai-Vy Nguyen' 'Fabrice Gamboa' 'Reda Chhaibi' 'Sixin Zhang'\n 'Serge Gratton' 'Thierry Giaccone']"
]
|
null | null | 2404.08480 | null | null | http://arxiv.org/pdf/2404.08480v1 | 2024-04-12T13:57:30Z | 2024-04-12T13:57:30Z | Decoding AI: The inside story of data analysis in ChatGPT | As a result of recent advancements in generative AI, the field of Data Science is prone to various changes. This review critically examines the Data Analysis (DA) capabilities of ChatGPT assessing its performance across a wide range of tasks. While DA provides researchers and practitioners with unprecedented analytical capabilities, it is far from being perfect, and it is important to recognize and address its limitations. | [
"['Ozan Evkaya' 'Miguel de Carvalho']"
]
|
null | null | 2404.08483 | null | null | http://arxiv.org/pdf/2404.08483v3 | 2024-05-22T16:36:29Z | 2024-04-12T14:03:41Z | Semantic Communication for Cooperative Multi-Task Processing over
Wireless Networks | In this paper, we have expanded the current status of semantic communication limited to processing one task to a more general system that can handle multiple tasks concurrently. In pursuit of this, we first introduced our definition of the "semantic source", enabling the interpretation of multiple semantics based on a single observation. A semantic encoder design is then introduced, featuring the division of the encoder into a common unit and multiple specific units enabling cooperative multi-task processing. Simulation results demonstrate the effectiveness of the proposed semantic source and the system design. Our approach employs information maximization (infomax) and end-to-end design principles. | [
"['Ahmad Halimi Razlighi' 'Carsten Bockelmann' 'Armin Dekorsy']"
]
|
null | null | 2404.08495 | null | null | http://arxiv.org/pdf/2404.08495v3 | 2024-04-16T17:36:39Z | 2024-04-12T14:25:49Z | Dataset Reset Policy Optimization for RLHF | Reinforcement Learning (RL) from Human Preference-based feedback is a popular paradigm for fine-tuning generative models, which has produced impressive models such as GPT-4 and Claude3 Opus. This framework often consists of two steps: learning a reward model from an offline preference dataset followed by running online RL to optimize the learned reward model. In this work, leveraging the idea of reset, we propose a new RLHF algorithm with provable guarantees. Motivated by the fact that offline preference dataset provides informative states (i.e., data that is preferred by the labelers), our new algorithm, Dataset Reset Policy Optimization (DR-PO), integrates the existing offline preference dataset into the online policy training procedure via dataset reset: it directly resets the policy optimizer to the states in the offline dataset, instead of always starting from the initial state distribution. In theory, we show that DR-PO learns to perform at least as good as any policy that is covered by the offline dataset under general function approximation with finite sample complexity. In experiments, we demonstrate that on both the TL;DR summarization and the Anthropic Helpful Harmful (HH) dataset, the generation from DR-PO is better than that from Proximal Policy Optimization (PPO) and Direction Preference Optimization (DPO), under the metric of GPT4 win-rate. Code for this work can be found at https://github.com/Cornell-RL/drpo. | [
"['Jonathan D. Chang' 'Wenhao Zhan' 'Owen Oertell' 'Kianté Brantley'\n 'Dipendra Misra' 'Jason D. Lee' 'Wen Sun']"
]
|
null | null | 2404.08509 | null | null | http://arxiv.org/pdf/2404.08509v1 | 2024-04-12T14:46:15Z | 2024-04-12T14:46:15Z | Efficient Interactive LLM Serving with Proxy Model-based Sequence Length
Prediction | Large language models (LLMs) have been driving a new wave of interactive AI applications across numerous domains. However, efficiently serving LLM inference requests is challenging due to their unpredictable execution times originating from the autoregressive nature of generative models. Existing LLM serving systems exploit first-come-first-serve (FCFS) scheduling, suffering from head-of-line blocking issues. To address the non-deterministic nature of LLMs and enable efficient interactive LLM serving, we present a speculative shortest-job-first (SSJF) scheduler that uses a light proxy model to predict LLM output sequence lengths. Our open-source SSJF implementation does not require changes to memory management or batching strategies. Evaluations on real-world datasets and production workload traces show that SSJF reduces average job completion times by 30.5-39.6% and increases throughput by 2.2-3.6x compared to FCFS schedulers, across no batching, dynamic batching, and continuous batching settings. | [
"['Haoran Qiu' 'Weichao Mao' 'Archit Patke' 'Shengkun Cui' 'Saurabh Jha'\n 'Chen Wang' 'Hubertus Franke' 'Zbigniew T. Kalbarczyk' 'Tamer Başar'\n 'Ravishankar K. Iyer']"
]
|
null | null | 2404.08513 | null | null | http://arxiv.org/pdf/2404.08513v1 | 2024-04-12T14:53:36Z | 2024-04-12T14:53:36Z | Adversarial Imitation Learning via Boosting | Adversarial imitation learning (AIL) has stood out as a dominant framework across various imitation learning (IL) applications, with Discriminator Actor Critic (DAC) (Kostrikov et al.,, 2019) demonstrating the effectiveness of off-policy learning algorithms in improving sample efficiency and scalability to higher-dimensional observations. Despite DAC's empirical success, the original AIL objective is on-policy and DAC's ad-hoc application of off-policy training does not guarantee successful imitation (Kostrikov et al., 2019; 2020). Follow-up work such as ValueDICE (Kostrikov et al., 2020) tackles this issue by deriving a fully off-policy AIL objective. Instead in this work, we develop a novel and principled AIL algorithm via the framework of boosting. Like boosting, our new algorithm, AILBoost, maintains an ensemble of properly weighted weak learners (i.e., policies) and trains a discriminator that witnesses the maximum discrepancy between the distributions of the ensemble and the expert policy. We maintain a weighted replay buffer to represent the state-action distribution induced by the ensemble, allowing us to train discriminators using the entire data collected so far. In the weighted replay buffer, the contribution of the data from older policies are properly discounted with the weight computed based on the boosting framework. Empirically, we evaluate our algorithm on both controller state-based and pixel-based environments from the DeepMind Control Suite. AILBoost outperforms DAC on both types of environments, demonstrating the benefit of properly weighting replay buffer data for off-policy training. On state-based environments, DAC outperforms ValueDICE and IQ-Learn (Gary et al., 2021), achieving competitive performance with as little as one expert trajectory. | [
"['Jonathan D. Chang' 'Dhruv Sreenivas' 'Yingbing Huang' 'Kianté Brantley'\n 'Wen Sun']"
]
|
null | null | 2404.08517 | null | null | http://arxiv.org/pdf/2404.08517v1 | 2024-04-12T14:55:16Z | 2024-04-12T14:55:16Z | Online Safety Analysis for LLMs: a Benchmark, an Assessment, and a Path
Forward | While Large Language Models (LLMs) have seen widespread applications across numerous fields, their limited interpretability poses concerns regarding their safe operations from multiple aspects, e.g., truthfulness, robustness, and fairness. Recent research has started developing quality assurance methods for LLMs, introducing techniques such as offline detector-based or uncertainty estimation methods. However, these approaches predominantly concentrate on post-generation analysis, leaving the online safety analysis for LLMs during the generation phase an unexplored area. To bridge this gap, we conduct in this work a comprehensive evaluation of the effectiveness of existing online safety analysis methods on LLMs. We begin with a pilot study that validates the feasibility of detecting unsafe outputs in the early generation process. Following this, we establish the first publicly available benchmark of online safety analysis for LLMs, including a broad spectrum of methods, models, tasks, datasets, and evaluation metrics. Utilizing this benchmark, we extensively analyze the performance of state-of-the-art online safety analysis methods on both open-source and closed-source LLMs. This analysis reveals the strengths and weaknesses of individual methods and offers valuable insights into selecting the most appropriate method based on specific application scenarios and task requirements. Furthermore, we also explore the potential of using hybridization methods, i.e., combining multiple methods to derive a collective safety conclusion, to enhance the efficacy of online safety analysis for LLMs. Our findings indicate a promising direction for the development of innovative and trustworthy quality assurance methodologies for LLMs, facilitating their reliable deployments across diverse domains. | [
"['Xuan Xie' 'Jiayang Song' 'Zhehua Zhou' 'Yuheng Huang' 'Da Song' 'Lei Ma']"
]
|
null | null | 2404.08522 | null | null | http://arxiv.org/pdf/2404.08522v1 | 2024-04-12T15:02:14Z | 2024-04-12T15:02:14Z | Fuxi-DA: A Generalized Deep Learning Data Assimilation Framework for
Assimilating Satellite Observations | Data assimilation (DA), as an indispensable component within contemporary Numerical Weather Prediction (NWP) systems, plays a crucial role in generating the analysis that significantly impacts forecast performance. Nevertheless, the development of an efficient DA system poses significant challenges, particularly in establishing intricate relationships between the background data and the vast amount of multi-source observation data within limited time windows in operational settings. To address these challenges, researchers design complex pre-processing methods for each observation type, leveraging approximate modeling and the power of super-computing clusters to expedite solutions. The emergence of deep learning (DL) models has been a game-changer, offering unified multi-modal modeling, enhanced nonlinear representation capabilities, and superior parallelization. These advantages have spurred efforts to integrate DL models into various domains of weather modeling. Remarkably, DL models have shown promise in matching, even surpassing, the forecast accuracy of leading operational NWP models worldwide. This success motivates the exploration of DL-based DA frameworks tailored for weather forecasting models. In this study, we introduces FuxiDA, a generalized DL-based DA framework for assimilating satellite observations. By assimilating data from Advanced Geosynchronous Radiation Imager (AGRI) aboard Fengyun-4B, FuXi-DA consistently mitigates analysis errors and significantly improves forecast performance. Furthermore, through a series of single-observation experiments, Fuxi-DA has been validated against established atmospheric physics, demonstrating its consistency and reliability. | [
"['Xiaoze Xu' 'Xiuyu Sun' 'Wei Han' 'Xiaohui Zhong' 'Lei Chen' 'Hao Li']"
]
|
null | null | 2404.08523 | null | null | http://arxiv.org/pdf/2404.08523v1 | 2024-04-12T15:10:57Z | 2024-04-12T15:10:57Z | Advancing Forest Fire Prevention: Deep Reinforcement Learning for
Effective Firebreak Placement | Over the past decades, the increase in both frequency and intensity of large-scale wildfires due to climate change has emerged as a significant natural threat. The pressing need to design resilient landscapes capable of withstanding such disasters has become paramount, requiring the development of advanced decision-support tools. Existing methodologies, including Mixed Integer Programming, Stochastic Optimization, and Network Theory, have proven effective but are hindered by computational demands, limiting their applicability. In response to this challenge, we propose using artificial intelligence techniques, specifically Deep Reinforcement Learning, to address the complex problem of firebreak placement in the landscape. We employ value-function based approaches like Deep Q-Learning, Double Deep Q-Learning, and Dueling Double Deep Q-Learning. Utilizing the Cell2Fire fire spread simulator combined with Convolutional Neural Networks, we have successfully implemented a computational agent capable of learning firebreak locations within a forest environment, achieving good results. Furthermore, we incorporate a pre-training loop, initially teaching our agent to mimic a heuristic-based algorithm and observe that it consistently exceeds the performance of these solutions. Our findings underscore the immense potential of Deep Reinforcement Learning for operational research challenges, especially in fire prevention. Our approach demonstrates convergence with highly favorable results in problem instances as large as 40 x 40 cells, marking a significant milestone in applying Reinforcement Learning to this critical issue. To the best of our knowledge, this study represents a pioneering effort in using Reinforcement Learning to address the aforementioned problem, offering promising perspectives in fire prevention and landscape management | [
"['Lucas Murray' 'Tatiana Castillo' 'Jaime Carrasco' 'Andrés Weintraub'\n 'Richard Weber' 'Isaac Martín de Diego' 'José Ramón González'\n 'Jordi García-Gonzalo']"
]
|
null | null | 2404.08535 | null | null | http://arxiv.org/pdf/2404.08535v1 | 2024-04-12T15:30:03Z | 2024-04-12T15:30:03Z | Generalized Contrastive Learning for Multi-Modal Retrieval and Ranking | Contrastive learning has gained widespread adoption for retrieval tasks due to its minimal requirement for manual annotations. However, popular contrastive frameworks typically learn from binary relevance, making them ineffective at incorporating direct fine-grained rankings. In this paper, we curate a large-scale dataset featuring detailed relevance scores for each query-document pair to facilitate future research and evaluation. Subsequently, we propose Generalized Contrastive Learning for Multi-Modal Retrieval and Ranking (GCL), which is designed to learn from fine-grained rankings beyond binary relevance scores. Our results show that GCL achieves a 94.5% increase in NDCG@10 for in-domain and 26.3 to 48.8% increases for cold-start evaluations, all relative to the CLIP baseline and involving ground truth rankings. | [
"['Tianyu Zhu' 'Myong Chol Jung' 'Jesse Clark']"
]
|
null | null | 2404.08555 | null | null | http://arxiv.org/pdf/2404.08555v2 | 2024-04-16T00:22:16Z | 2024-04-12T15:54:15Z | RLHF Deciphered: A Critical Analysis of Reinforcement Learning from
Human Feedback for LLMs | State-of-the-art large language models (LLMs) have become indispensable tools for various tasks. However, training LLMs to serve as effective assistants for humans requires careful consideration. A promising approach is reinforcement learning from human feedback (RLHF), which leverages human feedback to update the model in accordance with human preferences and mitigate issues like toxicity and hallucinations. Yet, an understanding of RLHF for LLMs is largely entangled with initial design choices that popularized the method and current research focuses on augmenting those choices rather than fundamentally improving the framework. In this paper, we analyze RLHF through the lens of reinforcement learning principles to develop an understanding of its fundamentals, dedicating substantial focus to the core component of RLHF -- the reward model. Our study investigates modeling choices, caveats of function approximation, and their implications on RLHF training algorithms, highlighting the underlying assumptions made about the expressivity of reward. Our analysis improves the understanding of the role of reward models and methods for their training, concurrently revealing limitations of the current methodology. We characterize these limitations, including incorrect generalization, model misspecification, and the sparsity of feedback, along with their impact on the performance of a language model. The discussion and analysis are substantiated by a categorical review of current literature, serving as a reference for researchers and practitioners to understand the challenges of RLHF and build upon existing efforts. | [
"['Shreyas Chaudhari' 'Pranjal Aggarwal' 'Vishvak Murahari'\n 'Tanmay Rajpurohit' 'Ashwin Kalyan' 'Karthik Narasimhan'\n 'Ameet Deshpande' 'Bruno Castro da Silva']"
]
|
null | null | 2404.08557 | null | null | http://arxiv.org/pdf/2404.08557v1 | 2024-04-12T15:54:48Z | 2024-04-12T15:54:48Z | Scalability in Building Component Data Annotation: Enhancing Facade
Material Classification with Synthetic Data | Computer vision models trained on Google Street View images can create material cadastres. However, current approaches need manually annotated datasets that are difficult to obtain and often have class imbalance. To address these challenges, this paper fine-tuned a Swin Transformer model on a synthetic dataset generated with DALL-E and compared the performance to a similar manually annotated dataset. Although manual annotation remains the gold standard, the synthetic dataset performance demonstrates a reasonable alternative. The findings will ease annotation needed to develop material cadastres, offering architects insights into opportunities for material reuse, thus contributing to the reduction of demolition waste. | [
"['Josie Harrison' 'Alexander Hollberg' 'Yinan Yu']"
]
|
null | null | 2404.08562 | null | null | http://arxiv.org/pdf/2404.08562v1 | 2024-04-03T22:07:50Z | 2024-04-03T22:07:50Z | Dynamic Neural Control Flow Execution: An Agent-Based Deep Equilibrium
Approach for Binary Vulnerability Detection | Software vulnerabilities are a challenge in cybersecurity. Manual security patches are often difficult and slow to be deployed, while new vulnerabilities are created. Binary code vulnerability detection is less studied and more complex compared to source code, and this has important practical implications. Deep learning has become an efficient and powerful tool in the security domain, where it provides end-to-end and accurate prediction. Modern deep learning approaches learn the program semantics through sequence and graph neural networks, using various intermediate representation of programs, such as abstract syntax trees (AST) or control flow graphs (CFG). Due to the complex nature of program execution, the output of an execution depends on the many program states and inputs. Also, a CFG generated from static analysis can be an overestimation of the true program flow. Moreover, the size of programs often does not allow a graph neural network with fixed layers to aggregate global information. To address these issues, we propose DeepEXE, an agent-based implicit neural network that mimics the execution path of a program. We use reinforcement learning to enhance the branching decision at every program state transition and create a dynamic environment to learn the dependency between a vulnerability and certain program states. An implicitly defined neural network enables nearly infinite state transitions until convergence, which captures the structural information at a higher level. The experiments are conducted on two semi-synthetic and two real-world datasets. We show that DeepEXE is an accurate and efficient method and outperforms the state-of-the-art vulnerability detection methods. | [
"['Litao Li' 'Steven H. H. Ding' 'Andrew Walenstein' 'Philippe Charland'\n 'Benjamin C. M. Fung']"
]
|
null | null | 2404.08564 | null | null | http://arxiv.org/pdf/2404.08564v1 | 2024-04-02T03:42:18Z | 2024-04-02T03:42:18Z | Federated Distillation: A Survey | Federated Learning (FL) seeks to train a model collaboratively without sharing private training data from individual clients. Despite its promise, FL encounters challenges such as high communication costs for large-scale models and the necessity for uniform model architectures across all clients and the server. These challenges severely restrict the practical applications of FL. To address these limitations, the integration of knowledge distillation (KD) into FL has been proposed, forming what is known as Federated Distillation (FD). FD enables more flexible knowledge transfer between clients and the server, surpassing the mere sharing of model parameters. By eliminating the need for identical model architectures across clients and the server, FD mitigates the communication costs associated with training large-scale models. This paper aims to offer a comprehensive overview of FD, highlighting its latest advancements. It delves into the fundamental principles underlying the design of FD frameworks, delineates FD approaches for tackling various challenges, and provides insights into the diverse applications of FD across different scenarios. | [
"['Lin Li' 'Jianping Gou' 'Baosheng Yu' 'Lan Du' 'Zhang Yiand Dacheng Tao']"
]
|
null | null | 2404.08566 | null | null | http://arxiv.org/pdf/2404.08566v1 | 2024-04-12T16:08:32Z | 2024-04-12T16:08:32Z | Mitigating Receiver Impact on Radio Frequency Fingerprint Identification
via Domain Adaptation | Radio Frequency Fingerprint Identification (RFFI), which exploits non-ideal hardware-induced unique distortion resident in the transmit signals to identify an emitter, is emerging as a means to enhance the security of communication systems. Recently, machine learning has achieved great success in developing state-of-the-art RFFI models. However, few works consider cross-receiver RFFI problems, where the RFFI model is trained and deployed on different receivers. Due to altered receiver characteristics, direct deployment of RFFI model on a new receiver leads to significant performance degradation. To address this issue, we formulate the cross-receiver RFFI as a model adaptation problem, which adapts the trained model to unlabeled signals from a new receiver. We first develop a theoretical generalization error bound for the adaptation model. Motivated by the bound, we propose a novel method to solve the cross-receiver RFFI problem, which includes domain alignment and adaptive pseudo-labeling. The former aims at finding a feature space where both domains exhibit similar distributions, effectively reducing the domain discrepancy. Meanwhile, the latter employs a dynamic pseudo-labeling scheme to implicitly transfer the label information from the labeled receiver to the new receiver. Experimental results indicate that the proposed method can effectively mitigate the receiver impact and improve the cross-receiver RFFI performance. | [
"['Liu Yang' 'Qiang Li' 'Xiaoyang Ren' 'Yi Fang' 'Shafei Wang']"
]
|
null | null | 2404.08570 | null | null | http://arxiv.org/pdf/2404.08570v1 | 2024-04-12T16:13:10Z | 2024-04-12T16:13:10Z | Enhancing Autonomous Vehicle Training with Language Model Integration
and Critical Scenario Generation | This paper introduces CRITICAL, a novel closed-loop framework for autonomous vehicle (AV) training and testing. CRITICAL stands out for its ability to generate diverse scenarios, focusing on critical driving situations that target specific learning and performance gaps identified in the Reinforcement Learning (RL) agent. The framework achieves this by integrating real-world traffic dynamics, driving behavior analysis, surrogate safety measures, and an optional Large Language Model (LLM) component. It is proven that the establishment of a closed feedback loop between the data generation pipeline and the training process can enhance the learning rate during training, elevate overall system performance, and augment safety resilience. Our evaluations, conducted using the Proximal Policy Optimization (PPO) and the HighwayEnv simulation environment, demonstrate noticeable performance improvements with the integration of critical case generation and LLM analysis, indicating CRITICAL's potential to improve the robustness of AV systems and streamline the generation of critical scenarios. This ultimately serves to hasten the development of AV agents, expand the general scope of RL training, and ameliorate validation efforts for AV safety. | [
"['Hanlin Tian' 'Kethan Reddy' 'Yuxiang Feng' 'Mohammed Quddus'\n 'Yiannis Demiris' 'Panagiotis Angeloudis']"
]
|
null | null | 2404.08573 | null | null | http://arxiv.org/pdf/2404.08573v2 | 2024-05-09T12:38:57Z | 2024-03-30T16:02:53Z | Going Forward-Forward in Distributed Deep Learning | We introduce a new approach in distributed deep learning, utilizing Geoffrey Hinton's Forward-Forward (FF) algorithm to speed up the training of neural networks in distributed computing environments. Unlike traditional methods that rely on forward and backward passes, the FF algorithm employs a dual forward pass strategy, significantly diverging from the conventional backpropagation process. This novel method aligns more closely with the human brain's processing mechanisms, potentially offering a more efficient and biologically plausible approach to neural network training. Our research explores different implementations of the FF algorithm in distributed settings, to explore its capacity for parallelization. While the original FF algorithm focused on its ability to match the performance of the backpropagation algorithm, the parallelism aims to reduce training times and resource consumption, thereby addressing the long training times associated with the training of deep neural networks. Our evaluation shows a 3.75 times speed up on MNIST dataset without compromising accuracy when training a four-layer network with four compute nodes. The integration of the FF algorithm into distributed deep learning represents a significant step forward in the field, potentially revolutionizing the way neural networks are trained in distributed environments. | [
"['Ege Aktemur' 'Ege Zorlutuna' 'Kaan Bilgili' 'Tacettin Emre Bok'\n 'Berrin Yanikoglu' 'Suha Orhun Mutluergil']"
]
|
null | null | 2404.08579 | null | null | http://arxiv.org/pdf/2404.08579v1 | 2024-04-12T16:23:41Z | 2024-04-12T16:23:41Z | Small Models Are (Still) Effective Cross-Domain Argument Extractors | Effective ontology transfer has been a major goal of recent work on event argument extraction (EAE). Two methods in particular -- question answering (QA) and template infilling (TI) -- have emerged as promising approaches to this problem. However, detailed explorations of these techniques' ability to actually enable this transfer are lacking. In this work, we provide such a study, exploring zero-shot transfer using both techniques on six major EAE datasets at both the sentence and document levels. Further, we challenge the growing reliance on LLMs for zero-shot extraction, showing that vastly smaller models trained on an appropriate source ontology can yield zero-shot performance superior to that of GPT-3.5 or GPT-4. | [
"['William Gantt' 'Aaron Steven White']"
]
|
null | null | 2404.08601 | null | null | http://arxiv.org/pdf/2404.08601v1 | 2024-04-12T16:55:08Z | 2024-04-12T16:55:08Z | Generating Synthetic Time Series Data for Cyber-Physical Systems | Data augmentation is an important facilitator of deep learning applications in the time series domain. A gap is identified in the literature, demonstrating sparse exploration of the transformer, the dominant sequence model, for data augmentation in time series. A architecture hybridizing several successful priors is put forth and tested using a powerful time domain similarity metric. Results suggest the challenge of this domain, and several valuable directions for future work. | [
"['Alexander Sommers' 'Somayeh Bakhtiari Ramezani' 'Logan Cummins'\n 'Sudip Mittal' 'Shahram Rahimi' 'Maria Seale' 'Joseph Jaboure']"
]
|
null | null | 2404.08602 | null | null | http://arxiv.org/pdf/2404.08602v2 | 2024-06-04T09:43:45Z | 2024-04-12T17:01:25Z | Sliding down the stairs: how correlated latent variables accelerate
learning with neural networks | Neural networks extract features from data using stochastic gradient descent (SGD). In particular, higher-order input cumulants (HOCs) are crucial for their performance. However, extracting information from the $p$th cumulant of $d$-dimensional inputs is computationally hard: the number of samples required to recover a single direction from an order-$p$ tensor (tensor PCA) using online SGD grows as $d^{p-1}$, which is prohibitive for high-dimensional inputs. This result raises the question of how neural networks extract relevant directions from the HOCs of their inputs efficiently. Here, we show that correlations between latent variables along the directions encoded in different input cumulants speed up learning from higher-order correlations. We show this effect analytically by deriving nearly sharp thresholds for the number of samples required by a single neuron to weakly-recover these directions using online SGD from a random start in high dimensions. Our analytical results are confirmed in simulations of two-layer neural networks and unveil a new mechanism for hierarchical learning in neural networks. | [
"['Lorenzo Bardone' 'Sebastian Goldt']"
]
|
null | null | 2404.08608 | null | null | http://arxiv.org/pdf/2404.08608v1 | 2024-04-12T17:14:58Z | 2024-04-12T17:14:58Z | Hyperbolic Delaunay Geometric Alignment | Hyperbolic machine learning is an emerging field aimed at representing data with a hierarchical structure. However, there is a lack of tools for evaluation and analysis of the resulting hyperbolic data representations. To this end, we propose Hyperbolic Delaunay Geometric Alignment (HyperDGA) -- a similarity score for comparing datasets in a hyperbolic space. The core idea is counting the edges of the hyperbolic Delaunay graph connecting datapoints across the given sets. We provide an empirical investigation on synthetic and real-life biological data and demonstrate that HyperDGA outperforms the hyperbolic version of classical distances between sets. Furthermore, we showcase the potential of HyperDGA for evaluating latent representations inferred by a Hyperbolic Variational Auto-Encoder. | [
"['Aniss Aiman Medbouhi' 'Giovanni Luca Marchetti' 'Vladislav Polianskii'\n 'Alexander Kravberg' 'Petra Poklukar' 'Anastasia Varava' 'Danica Kragic']"
]
|
null | null | 2404.08624 | null | null | http://arxiv.org/pdf/2404.08624v1 | 2024-04-12T17:37:42Z | 2024-04-12T17:37:42Z | Regularized Gradient Clipping Provably Trains Wide and Deep Neural
Networks | In this work, we instantiate a regularized form of the gradient clipping algorithm and prove that it can converge to the global minima of deep neural network loss functions provided that the net is of sufficient width. We present empirical evidence that our theoretically founded regularized gradient clipping algorithm is also competitive with the state-of-the-art deep-learning heuristics. Hence the algorithm presented here constitutes a new approach to rigorous deep learning. The modification we do to standard gradient clipping is designed to leverage the PL* condition, a variant of the Polyak-Lojasiewicz inequality which was recently proven to be true for various neural networks for any depth within a neighborhood of the initialisation. | [
"['Matteo Tucat' 'Anirbit Mukherjee']"
]
|
null | null | 2404.08627 | null | null | http://arxiv.org/pdf/2404.08627v1 | 2024-04-12T17:41:05Z | 2024-04-12T17:41:05Z | Is ChatGPT Transforming Academics' Writing Style? | Based on one million arXiv papers submitted from May 2018 to January 2024, we assess the textual density of ChatGPT's writing style in their abstracts by means of a statistical analysis of word frequency changes. Our model is calibrated and validated on a mixture of real abstracts and ChatGPT-modified abstracts (simulated data) after a careful noise analysis. We find that ChatGPT is having an increasing impact on arXiv abstracts, especially in the field of computer science, where the fraction of ChatGPT-revised abstracts is estimated to be approximately 35%, if we take the output of one of the simplest prompts, "revise the following sentences", as a baseline. We conclude with an analysis of both positive and negative aspects of the penetration of ChatGPT into academics' writing style. | [
"['Mingmeng Geng' 'Roberto Trotta']"
]
|
null | null | 2404.08634 | null | null | http://arxiv.org/pdf/2404.08634v1 | 2024-04-12T17:53:34Z | 2024-04-12T17:53:34Z | Pre-training Small Base LMs with Fewer Tokens | We study the effectiveness of a simple approach to develop a small base language model (LM) starting from an existing large base LM: first inherit a few transformer blocks from the larger LM, and then train this smaller model on a very small subset (0.1%) of the raw pretraining data of the larger model. We call our simple recipe Inheritune and first demonstrate it for building a small base LM with 1.5B parameters using 1B tokens (and a starting few layers of larger LM of 3B parameters); we do this using a single A6000 GPU for less than half a day. Across 9 diverse evaluation datasets as well as the MMLU benchmark, the resulting model compares favorably to publicly available base models of 1B-2B size, some of which have been trained using 50-1000 times more tokens. We investigate Inheritune in a slightly different setting where we train small LMs utilizing larger LMs and their full pre-training dataset. Here we show that smaller LMs trained utilizing some of the layers of GPT2-medium (355M) and GPT-2-large (770M) can effectively match the val loss of their bigger counterparts when trained from scratch for the same number of training steps on OpenWebText dataset with 9B tokens. We analyze our recipe with extensive experiments and demonstrate it efficacy on diverse settings. Our code is available at https://github.com/sanyalsunny111/LLM-Inheritune. | [
"['Sunny Sanyal' 'Sujay Sanghavi' 'Alexandros G. Dimakis']"
]
|
null | null | 2404.08652 | null | null | http://arxiv.org/pdf/2404.08652v1 | 2024-03-19T05:42:29Z | 2024-03-19T05:42:29Z | Algorithm for AGC index management against crowded radio environment | This paper describes a receiver that uses an innovative method to predict, according to history of receiver operating metrics (packet lost/well received), the optimum automatic gain control (AGC) index or most appropriate variable gain range to be used for next packet reception, anticipating an interferer appearing during the payload reception. This allows the receiver to have higher immunity to interferers even if they occur during the gain frozen payload reception period whilst still ensuring an optimum sensitivity level. As a result, the method allows setting the receiver gain to get an optimum trade-off between reception sensitivity and random interferer immunity. | [
"['Morgane Joly' 'Fabian Rivière' 'Éric Renault']"
]
|
null | null | 2404.08655 | null | null | http://arxiv.org/pdf/2404.08655v1 | 2024-03-24T21:44:14Z | 2024-03-24T21:44:14Z | Transformer-based Joint Modelling for Automatic Essay Scoring and
Off-Topic Detection | Automated Essay Scoring (AES) systems are widely popular in the market as they constitute a cost-effective and time-effective option for grading systems. Nevertheless, many studies have demonstrated that the AES system fails to assign lower grades to irrelevant responses. Thus, detecting the off-topic response in automated essay scoring is crucial in practical tasks where candidates write unrelated text responses to the given task in the question. In this paper, we are proposing an unsupervised technique that jointly scores essays and detects off-topic essays. The proposed Automated Open Essay Scoring (AOES) model uses a novel topic regularization module (TRM), which can be attached on top of a transformer model, and is trained using a proposed hybrid loss function. After training, the AOES model is further used to calculate the Mahalanobis distance score for off-topic essay detection. Our proposed method outperforms the baseline we created and earlier conventional methods on two essay-scoring datasets in off-topic detection as well as on-topic scoring. Experimental evaluation results on different adversarial strategies also show how the suggested method is robust for detecting possible human-level perturbations. | [
"['Sourya Dipta Das' 'Yash Vadi' 'Kuldeep Yadav']"
]
|
null | null | 2404.08657 | null | null | http://arxiv.org/pdf/2404.08657v1 | 2024-03-25T09:30:19Z | 2024-03-25T09:30:19Z | Advancing Extrapolative Predictions of Material Properties through
Learning to Learn | Recent advancements in machine learning have showcased its potential to significantly accelerate the discovery of new materials. Central to this progress is the development of rapidly computable property predictors, enabling the identification of novel materials with desired properties from vast material spaces. However, the limited availability of data resources poses a significant challenge in data-driven materials research, particularly hindering the exploration of innovative materials beyond the boundaries of existing data. While machine learning predictors are inherently interpolative, establishing a general methodology to create an extrapolative predictor remains a fundamental challenge, limiting the search for innovative materials beyond existing data boundaries. In this study, we leverage an attention-based architecture of neural networks and meta-learning algorithms to acquire extrapolative generalization capability. The meta-learners, experienced repeatedly with arbitrarily generated extrapolative tasks, can acquire outstanding generalization capability in unexplored material spaces. Through the tasks of predicting the physical properties of polymeric materials and hybrid organic--inorganic perovskites, we highlight the potential of such extrapolatively trained models, particularly with their ability to rapidly adapt to unseen material domains in transfer learning scenarios. | [
"['Kohei Noda' 'Araki Wakiuchi' 'Yoshihiro Hayashi' 'Ryo Yoshida']"
]
|
null | null | 2404.08660 | null | null | http://arxiv.org/pdf/2404.08660v1 | 2024-03-27T18:53:04Z | 2024-03-27T18:53:04Z | How Does Message Passing Improve Collaborative Filtering? | Collaborative filtering (CF) has exhibited prominent results for recommender systems and been broadly utilized for real-world applications. A branch of research enhances CF methods by message passing used in graph neural networks, due to its strong capabilities of extracting knowledge from graph-structured data, like user-item bipartite graphs that naturally exist in CF. They assume that message passing helps CF methods in a manner akin to its benefits for graph-based learning tasks in general. However, even though message passing empirically improves CF, whether or not this assumption is correct still needs verification. To address this gap, we formally investigate why message passing helps CF from multiple perspectives and show that many assumptions made by previous works are not entirely accurate. With our curated ablation studies and theoretical analyses, we discover that (1) message passing improves the CF performance primarily by additional representations passed from neighbors during the forward pass instead of additional gradient updates to neighbor representations during the model back-propagation and (ii) message passing usually helps low-degree nodes more than high-degree nodes. Utilizing these novel findings, we present Test-time Aggregation for CF, namely TAG-CF, a test-time augmentation framework that only conducts message passing once at inference time. The key novelty of TAG-CF is that it effectively utilizes graph knowledge while circumventing most of notorious computational overheads of message passing. Besides, TAG-CF is extremely versatile can be used as a plug-and-play module to enhance representations trained by different CF supervision signals. Evaluated on six datasets, TAG-CF consistently improves the recommendation performance of CF methods without graph by up to 39.2% on cold users and 31.7% on all users, with little to no extra computational overheads. | [
"['Mingxuan Ju' 'William Shiao' 'Zhichun Guo' 'Yanfang Ye' 'Yozen Liu'\n 'Neil Shah' 'Tong Zhao']"
]
|
null | null | 2404.08662 | null | null | http://arxiv.org/pdf/2404.08662v1 | 2024-03-28T09:59:59Z | 2024-03-28T09:59:59Z | FewUser: Few-Shot Social User Geolocation via Contrastive Learning | To address the challenges of scarcity in geotagged data for social user geolocation, we propose FewUser, a novel framework for Few-shot social User geolocation. We incorporate a contrastive learning strategy between users and locations to improve geolocation performance with no or limited training data. FewUser features a user representation module that harnesses a pre-trained language model (PLM) and a user encoder to process and fuse diverse social media inputs effectively. To bridge the gap between PLM's knowledge and geographical data, we introduce a geographical prompting module with hard, soft, and semi-soft prompts, to enhance the encoding of location information. Contrastive learning is implemented through a contrastive loss and a matching loss, complemented by a hard negative mining strategy to refine the learning process. We construct two datasets TwiU and FliU, containing richer metadata than existing benchmarks, to evaluate FewUser and the extensive experiments demonstrate that FewUser significantly outperforms state-of-the-art methods in both zero-shot and various few-shot settings, achieving absolute improvements of 26.95% and textbf{41.62%} on TwiU and FliU, respectively, with only one training sample per class. We further conduct a comprehensive analysis to investigate the impact of user representation on geolocation performance and the effectiveness of FewUser's components, offering valuable insights for future research in this area. | [
"['Menglin Li' 'Kwan Hui Lim']"
]
|
null | null | 2404.08664 | null | null | http://arxiv.org/abs/2404.08664v1 | 2024-03-29T13:15:46Z | 2024-03-29T13:15:46Z | Identifying Banking Transaction Descriptions via Support Vector Machine
Short-Text Classification Based on a Specialized Labelled Corpus | Short texts are omnipresent in real-time news, social network commentaries, etc. Traditional text representation methods have been successfully applied to self-contained documents of medium size. However, information in short texts is often insufficient, due, for example, to the use of mnemonics, which makes them hard to classify. Therefore, the particularities of specific domains must be exploited. In this article we describe a novel system that combines Natural Language Processing techniques with Machine Learning algorithms to classify banking transaction descriptions for personal finance management, a problem that was not previously considered in the literature. We trained and tested that system on a labelled dataset with real customer transactions that will be available to other researchers on request. Motivated by existing solutions in spam detection, we also propose a short text similarity detector to reduce training set size based on the Jaccard distance. Experimental results with a two-stage classifier combining this detector with a SVM indicate a high accuracy in comparison with alternative approaches, taking into account complexity and computing time. Finally, we present a use case with a personal finance application, CoinScrap, which is available at Google Play and App Store. | [
"['Silvia García-Méndez' 'Milagros Fernández-Gavilanes'\n 'Jonathan Juncal-Martínez' 'Francisco J. González-Castaño'\n 'Oscar Barba Seara']"
]
|
null | null | 2404.08665 | null | null | http://arxiv.org/abs/2404.08665v1 | 2024-03-30T16:46:25Z | 2024-03-30T16:46:25Z | Targeted aspect-based emotion analysis to detect opportunities and
precaution in financial Twitter messages | Microblogging platforms, of which Twitter is a representative example, are valuable information sources for market screening and financial models. In them, users voluntarily provide relevant information, including educated knowledge on investments, reacting to the state of the stock markets in real-time and, often, influencing this state. We are interested in the user forecasts in financial, social media messages expressing opportunities and precautions about assets. We propose a novel Targeted Aspect-Based Emotion Analysis (TABEA) system that can individually discern the financial emotions (positive and negative forecasts) on the different stock market assets in the same tweet (instead of making an overall guess about that whole tweet). It is based on Natural Language Processing (NLP) techniques and Machine Learning streaming algorithms. The system comprises a constituency parsing module for parsing the tweets and splitting them into simpler declarative clauses; an offline data processing module to engineer textual, numerical and categorical features and analyse and select them based on their relevance; and a stream classification module to continuously process tweets on-the-fly. Experimental results on a labelled data set endorse our solution. It achieves over 90% precision for the target emotions, financial opportunity, and precaution on Twitter. To the best of our knowledge, no prior work in the literature has addressed this problem despite its practical interest in decision-making, and we are not aware of any previous NLP nor online Machine Learning approaches to TABEA. | [
"['Silvia García-Méndez' 'Francisco de Arriba-Pérez' 'Ana Barros-Vila'\n 'Francisco J. González-Castaño']"
]
|
null | null | 2404.08666 | null | null | http://arxiv.org/pdf/2404.08666v2 | 2024-07-15T14:07:16Z | 2024-03-31T15:13:15Z | Revealing Trends in Datasets from the 2022 ACL and EMNLP Conferences | Natural language processing (NLP) has grown significantly since the advent of the Transformer architecture. Transformers have given birth to pre-trained large language models (PLMs). There has been tremendous improvement in the performance of NLP systems across several tasks. NLP systems are on par or, in some cases, better than humans at accomplishing specific tasks. However, it remains the norm that emph{better quality datasets at the time of pretraining enable PLMs to achieve better performance, regardless of the task.} The need to have quality datasets has prompted NLP researchers to continue creating new datasets to satisfy particular needs. For example, the two top NLP conferences, ACL and EMNLP, accepted ninety-two papers in 2022, introducing new datasets. This work aims to uncover the trends and insights mined within these datasets. Moreover, we provide valuable suggestions to researchers interested in curating datasets in the future. | [
"['Jesse Atuhurra' 'Hidetaka Kamigaito']"
]
|
null | null | 2404.08670 | null | null | http://arxiv.org/pdf/2404.08670v1 | 2024-04-03T04:25:07Z | 2024-04-03T04:25:07Z | A Bayesian Regression Approach for Estimating the Impact of COVID-19 on
Consumer Behavior in the Restaurant Industry | The COVID-19 pandemic has had a long-term impact on industries worldwide, with the hospitality and food industry facing significant challenges, leading to the permanent closure of many restaurants and the loss of jobs. In this study, we developed an innovative analytical framework using Hamiltonian Monte Carlo for predictive modeling with Bayesian regression, aiming to estimate the change point in consumer behavior towards different types of restaurants due to COVID-19. Our approach emphasizes a novel method in computational analysis, providing insights into customer behavior changes before and after the pandemic. This research contributes to understanding the effects of COVID-19 on the restaurant industry and is valuable for restaurant owners and policymakers. | [
"['H. Hinduja' 'N. Mandal']"
]
|
null | null | 2404.08671 | null | null | http://arxiv.org/pdf/2404.08671v1 | 2024-04-03T17:15:45Z | 2024-04-03T17:15:45Z | Navigating the Evaluation Funnel to Optimize Iteration Speed for
Recommender Systems | Over the last decades has emerged a rich literature on the evaluation of recommendation systems. However, less is written about how to efficiently combine different evaluation methods from this rich field into a single efficient evaluation funnel. In this paper we aim to build intuition for how to choose evaluation methods, by presenting a novel framework that simplifies the reasoning around the evaluation funnel for a recommendation system. Our contribution is twofold. First we present our framework for how to decompose the definition of success to construct efficient evaluation funnels, focusing on how to identify and discard non-successful iterations quickly. We show that decomposing the definition of success into smaller necessary criteria for success enables early identification of non-successful ideas. Second, we give an overview of the most common and useful evaluation methods, discuss their pros and cons, and how they fit into, and complement each other in, the evaluation process. We go through so-called offline and online evaluation methods such as counterfactual logging, validation, verification, A/B testing, and interleaving. The paper concludes with some general discussion and advice on how to design an efficient evaluation process for recommender systems. | [
"['Claire Schultzberg' 'Brammert Ottens']"
]
|
null | null | 2404.08672 | null | null | http://arxiv.org/pdf/2404.08672v1 | 2024-04-05T05:14:46Z | 2024-04-05T05:14:46Z | Taxonomy and Analysis of Sensitive User Queries in Generative AI Search | Although there has been a growing interest among industries to integrate generative LLMs into their services, limited experiences and scarcity of resources acts as a barrier in launching and servicing large-scale LLM-based conversational services. In this paper, we share our experiences in developing and operating generative AI models within a national-scale search engine, with a specific focus on the sensitiveness of user queries. We propose a taxonomy for sensitive search queries, outline our approaches, and present a comprehensive analysis report on sensitive queries from actual users. | [
"['Hwiyeol Jo' 'Taiwoo Park' 'Nayoung Choi' 'Changbong Kim' 'Ohjoon Kwon'\n 'Donghyeon Jeon' 'Hyunwoo Lee' 'Eui-Hyeon Lee' 'Kyoungho Shin'\n 'Sun Suk Lim' 'Kyungmi Kim' 'Jihye Lee' 'Sun Kim']"
]
|
null | null | 2404.08673 | null | null | http://arxiv.org/pdf/2404.08673v1 | 2024-04-05T16:14:36Z | 2024-04-05T16:14:36Z | Sentiment analysis and random forest to classify LLM versus human source
applied to Scientific Texts | After the launch of ChatGPT v.4 there has been a global vivid discussion on the ability of this artificial intelligence powered platform and some other similar ones for the automatic production of all kinds of texts, including scientific and technical texts. This has triggered a reflection in many institutions on whether education and academic procedures should be adapted to the fact that in future many texts we read will not be written by humans (students, scholars, etc.), at least, not entirely. In this work it is proposed a new methodology to classify texts coming from an automatic text production engine or a human, based on Sentiment Analysis as a source for feature engineering independent variables and then train with them a Random Forest classification algorithm. Using four different sentiment lexicons, a number of new features where produced, and then fed to a machine learning random forest methodology, to train such a model. Results seem very convincing that this may be a promising research line to detect fraud, in such environments where human are supposed to be the source of texts. | [
"['Javier J. Sanchez-Medina']"
]
|
null | null | 2404.08676 | null | null | http://arxiv.org/pdf/2404.08676v3 | 2024-06-24T08:50:22Z | 2024-04-06T15:01:47Z | ALERT: A Comprehensive Benchmark for Assessing Large Language Models'
Safety through Red Teaming | When building Large Language Models (LLMs), it is paramount to bear safety in mind and protect them with guardrails. Indeed, LLMs should never generate content promoting or normalizing harmful, illegal, or unethical behavior that may contribute to harm to individuals or society. This principle applies to both normal and adversarial use. In response, we introduce ALERT, a large-scale benchmark to assess safety based on a novel fine-grained risk taxonomy. It is designed to evaluate the safety of LLMs through red teaming methodologies and consists of more than 45k instructions categorized using our novel taxonomy. By subjecting LLMs to adversarial testing scenarios, ALERT aims to identify vulnerabilities, inform improvements, and enhance the overall safety of the language models. Furthermore, the fine-grained taxonomy enables researchers to perform an in-depth evaluation that also helps one to assess the alignment with various policies. In our experiments, we extensively evaluate 10 popular open- and closed-source LLMs and demonstrate that many of them still struggle to attain reasonable levels of safety. | [
"['Simone Tedeschi' 'Felix Friedrich' 'Patrick Schramowski'\n 'Kristian Kersting' 'Roberto Navigli' 'Huu Nguyen' 'Bo Li']"
]
|
null | null | 2404.08679 | null | null | http://arxiv.org/pdf/2404.08679v1 | 2024-04-07T10:32:49Z | 2024-04-07T10:32:49Z | Your Finetuned Large Language Model is Already a Powerful
Out-of-distribution Detector | We revisit the likelihood ratio between a pretrained large language model (LLM) and its finetuned variant as a criterion for out-of-distribution (OOD) detection. The intuition behind such a criterion is that, the pretrained LLM has the prior knowledge about OOD data due to its large amount of training data, and once finetuned with the in-distribution data, the LLM has sufficient knowledge to distinguish their difference. Leveraging the power of LLMs, we show that, for the first time, the likelihood ratio can serve as an effective OOD detector. Moreover, we apply the proposed LLM-based likelihood ratio to detect OOD questions in question-answering (QA) systems, which can be used to improve the performance of specialized LLMs for general questions. Given that likelihood can be easily obtained by the loss functions within contemporary neural network frameworks, it is straightforward to implement this approach in practice. Since both the pretrained LLMs and its various finetuned models are available, our proposed criterion can be effortlessly incorporated for OOD detection without the need for further training. We conduct comprehensive evaluation across on multiple settings, including far OOD, near OOD, spam detection, and QA scenarios, to demonstrate the effectiveness of the method. | [
"['Andi Zhang' 'Tim Z. Xiao' 'Weiyang Liu' 'Robert Bamler' 'Damon Wischik']"
]
|
null | null | 2404.08683 | null | null | http://arxiv.org/pdf/2404.08683v1 | 2024-04-08T16:18:33Z | 2024-04-08T16:18:33Z | Text clustering applied to data augmentation in legal contexts | Data analysis and machine learning are of preeminent importance in the legal domain, especially in tasks like clustering and text classification. In this study, we harnessed the power of natural language processing tools to enhance datasets meticulously curated by experts. This process significantly improved the classification workflow for legal texts using machine learning techniques. We considered the Sustainable Development Goals (SDGs) data from the United Nations 2030 Agenda as a practical case study. Data augmentation clustering-based strategy led to remarkable enhancements in the accuracy and sensitivity metrics of classification models. For certain SDGs within the 2030 Agenda, we observed performance gains of over 15%. In some cases, the example base expanded by a noteworthy factor of 5. When dealing with unclassified legal texts, data augmentation strategies centered around clustering prove to be highly effective. They provide a valuable means to expand the existing knowledge base without the need for labor-intensive manual classification efforts. | [
"['Lucas José Gonçalves Freitas' 'Thaís Rodrigues' 'Guilherme Rodrigues'\n 'Pamella Edokawa' 'Ariane Farias']"
]
|
null | null | 2404.08685 | null | null | http://arxiv.org/pdf/2404.08685v1 | 2024-04-08T18:33:59Z | 2024-04-08T18:33:59Z | Neural Sequence-to-Sequence Modeling with Attention by Leveraging Deep
Learning Architectures for Enhanced Contextual Understanding in Abstractive
Text Summarization | Automatic text summarization (TS) plays a pivotal role in condensing large volumes of information into concise, coherent summaries, facilitating efficient information retrieval and comprehension. This paper presents a novel framework for abstractive TS of single documents, which integrates three dominant aspects: structural, semantic, and neural-based approaches. The proposed framework merges machine learning and knowledge-based techniques to achieve a unified methodology. The framework consists of three main phases: pre-processing, machine learning, and post-processing. In the pre-processing phase, a knowledge-based Word Sense Disambiguation (WSD) technique is employed to generalize ambiguous words, enhancing content generalization. Semantic content generalization is then performed to address out-of-vocabulary (OOV) or rare words, ensuring comprehensive coverage of the input document. Subsequently, the generalized text is transformed into a continuous vector space using neural language processing techniques. A deep sequence-to-sequence (seq2seq) model with an attention mechanism is employed to predict a generalized summary based on the vector representation. In the post-processing phase, heuristic algorithms and text similarity metrics are utilized to refine the generated summary further. Concepts from the generalized summary are matched with specific entities, enhancing coherence and readability. Experimental evaluations conducted on prominent datasets, including Gigaword, Duc 2004, and CNN/DailyMail, demonstrate the effectiveness of the proposed framework. Results indicate significant improvements in handling rare and OOV words, outperforming existing state-of-the-art deep learning techniques. The proposed framework presents a comprehensive and unified approach towards abstractive TS, combining the strengths of structure, semantics, and neural-based methodologies. | [
"['Bhavith Chandra Challagundla' 'Chakradhar Peddavenkatagari']"
]
|
null | null | 2404.08686 | null | null | http://arxiv.org/pdf/2404.08686v1 | 2024-04-09T04:54:08Z | 2024-04-09T04:54:08Z | Extractive text summarisation of Privacy Policy documents using machine
learning approaches | This work demonstrates two Privacy Policy (PP) summarisation models based on two different clustering algorithms: K-means clustering and Pre-determined Centroid (PDC) clustering. K-means is decided to be used for the first model after an extensive evaluation of ten commonly used clustering algorithms. The summariser model based on the PDC-clustering algorithm summarises PP documents by segregating individual sentences by Euclidean distance from each sentence to the pre-defined cluster centres. The cluster centres are defined according to General Data Protection Regulation (GDPR)'s 14 essential topics that must be included in any privacy notices. The PDC model outperformed the K-means model for two evaluation methods, Sum of Squared Distance (SSD) and ROUGE by some margin (27% and 24% respectively). This result contrasts the K-means model's better performance in the general clustering of sentence vectors before running the task-specific evaluation. This indicates the effectiveness of operating task-specific fine-tuning measures on unsupervised machine-learning models. The summarisation mechanisms implemented in this paper demonstrates an idea of how to efficiently extract essential sentences that should be included in any PP documents. The summariser models could be further developed to an application that tests the GDPR-compliance (or any data privacy legislation) of PP documents. | [
"['Chanwoo Choi']"
]
|
null | null | 2404.08690 | null | null | http://arxiv.org/pdf/2404.08690v1 | 2024-04-09T22:56:05Z | 2024-04-09T22:56:05Z | Towards Building a Robust Toxicity Predictor | Recent NLP literature pays little attention to the robustness of toxicity language predictors, while these systems are most likely to be used in adversarial contexts. This paper presents a novel adversarial attack, texttt{ToxicTrap}, introducing small word-level perturbations to fool SOTA text classifiers to predict toxic text samples as benign. ToxicTrap exploits greedy based search strategies to enable fast and effective generation of toxic adversarial examples. Two novel goal function designs allow ToxicTrap to identify weaknesses in both multiclass and multilabel toxic language detectors. Our empirical results show that SOTA toxicity text classifiers are indeed vulnerable to the proposed attacks, attaining over 98% attack success rates in multilabel cases. We also show how a vanilla adversarial training and its improved version can help increase robustness of a toxicity detector even against unseen attacks. | [
"['Dmitriy Bespalov' 'Sourav Bhabesh' 'Yi Xiang' 'Liutong Zhou' 'Yanjun Qi']"
]
|
null | null | 2404.08698 | null | null | http://arxiv.org/pdf/2404.08698v2 | 2024-07-10T07:38:32Z | 2024-04-10T16:11:09Z | Lossless Acceleration of Large Language Model via Adaptive N-gram
Parallel Decoding | While Large Language Models (LLMs) have shown remarkable abilities, they are hindered by significant resource consumption and considerable latency due to autoregressive processing. In this study, we introduce Adaptive N-gram Parallel Decoding (ANPD), an innovative and lossless approach that accelerates inference by allowing the simultaneous generation of multiple tokens. ANPD incorporates a two-stage approach: it begins with a rapid drafting phase that employs an N-gram module, which adapts based on the current interactive context, followed by a verification phase, during which the original LLM assesses and confirms the proposed tokens. Consequently, ANPD preserves the integrity of the LLM's original output while enhancing processing speed. We further leverage a multi-level architecture for the N-gram module to enhance the precision of the initial draft, consequently reducing inference latency. ANPD eliminates the need for retraining or extra GPU memory, making it an efficient and plug-and-play enhancement. In our experiments, models such as LLaMA and its fine-tuned variants have shown speed improvements up to 3.67x, validating the effectiveness of our proposed ANPD. | [
"['Jie Ou' 'Yueming Chen' 'Wenhong Tian']"
]
|
null | null | 2404.08699 | null | null | http://arxiv.org/pdf/2404.08699v2 | 2024-04-21T23:51:29Z | 2024-04-10T16:30:09Z | Analyzing the Impact of Data Selection and Fine-Tuning on Economic and
Political Biases in LLMs | In an era where language models are increasingly integrated into decision-making and communication, understanding the biases within Large Language Models (LLMs) becomes imperative, especially when these models are applied in the economic and political domains. This work investigates the impact of fine-tuning and data selection on economic and political biases in LLM. We explore the methodological aspects of biasing LLMs towards specific ideologies, mindful of the biases that arise from their extensive training on diverse datasets. Our approach, distinct from earlier efforts that either focus on smaller models or entail resource-intensive pre-training, employs Parameter-Efficient Fine-Tuning (PEFT) techniques. These techniques allow for the alignment of LLMs with targeted ideologies by modifying a small subset of parameters. We introduce a systematic method for dataset selection, annotation, and instruction tuning, and we assess its effectiveness through both quantitative and qualitative evaluations. Our work analyzes the potential of embedding specific biases into LLMs and contributes to the dialogue on the ethical application of AI, highlighting the importance of deploying AI in a manner that aligns with societal values. | [
"['Ahmed Agiza' 'Mohamed Mostagir' 'Sherief Reda']"
]
|
null | null | 2404.08701 | null | null | http://arxiv.org/pdf/2404.08701v1 | 2024-04-11T01:16:33Z | 2024-04-11T01:16:33Z | Can Contrastive Learning Refine Embeddings | Recent advancements in contrastive learning have revolutionized self-supervised representation learning and achieved state-of-the-art performance on benchmark tasks. While most existing methods focus on applying contrastive learning to input data modalities such as images, natural language sentences, or networks, they overlook the potential of utilizing outputs from previously trained encoders. In this paper, we introduce SIMSKIP, a novel contrastive learning framework that specifically refines input embeddings for downstream tasks. Unlike traditional unsupervised learning approaches, SIMSKIP takes advantage of the output embeddings of encoder models as its input. Through theoretical analysis, we provide evidence that applying SIMSKIP does not result in larger upper bounds on downstream task errors than those of the original embeddings, which serve as SIMSKIP's input. Experimental results on various open datasets demonstrate that the embeddings produced by SIMSKIP improve performance on downstream tasks. | [
"['Lihui Liu' 'Jinha Kim' 'Vidit Bansal']"
]
|
null | null | 2404.08702 | null | null | http://arxiv.org/pdf/2404.08702v1 | 2024-04-11T05:03:40Z | 2024-04-11T05:03:40Z | Predictive Modelling of Air Quality Index (AQI) Across Diverse Cities
and States of India using Machine Learning: Investigating the Influence of
Punjab's Stubble Burning on AQI Variability | Air pollution is a common and serious problem nowadays and it cannot be ignored as it has harmful impacts on human health. To address this issue proactively, people should be aware of their surroundings, which means the environment where they survive. With this motive, this research has predicted the AQI based on different air pollutant concentrations in the atmosphere. The dataset used for this research has been taken from the official website of CPCB. The dataset has the air pollutant concentration from 22 different monitoring stations in different cities of Delhi, Haryana, and Punjab. This data is checked for null values and outliers. But, the most important thing to note is the correct understanding and imputation of such values rather than ignoring or doing wrong imputation. The time series data has been used in this research which is tested for stationarity using The Dickey-Fuller test. Further different ML models like CatBoost, XGBoost, Random Forest, SVM regressor, time series model SARIMAX, and deep learning model LSTM have been used to predict AQI. For the performance evaluation of different models, I used MSE, RMSE, MAE, and R2. It is observed that Random Forest performed better as compared to other models. | [
"['Kamaljeet Kaur Sidhu' 'Habeeb Balogun' 'Kazeem Oluwakemi Oseni']"
]
|
null | null | 2404.08705 | null | null | http://arxiv.org/pdf/2404.08705v1 | 2024-04-11T07:39:22Z | 2024-04-11T07:39:22Z | Introducing L2M3, A Multilingual Medical Large Language Model to Advance
Health Equity in Low-Resource Regions | Addressing the imminent shortfall of 10 million health workers by 2030, predominantly in Low- and Middle-Income Countries (LMICs), this paper introduces an innovative approach that harnesses the power of Large Language Models (LLMs) integrated with machine translation models. This solution is engineered to meet the unique needs of Community Health Workers (CHWs), overcoming language barriers, cultural sensitivities, and the limited availability of medical dialog datasets. I have crafted a model that not only boasts superior translation capabilities but also undergoes rigorous fine-tuning on open-source datasets to ensure medical accuracy and is equipped with comprehensive safety features to counteract the risks of misinformation. Featuring a modular design, this approach is specifically structured for swift adaptation across various linguistic and cultural contexts, utilizing open-source components to significantly reduce healthcare operational costs. This strategic innovation markedly improves the accessibility and quality of healthcare services by providing CHWs with contextually appropriate medical knowledge and diagnostic tools. This paper highlights the transformative impact of this context-aware LLM, underscoring its crucial role in addressing the global healthcare workforce deficit and propelling forward healthcare outcomes in LMICs. | [
"['Agasthya Gangavarapu']"
]
|
null | null | 2404.08707 | null | null | http://arxiv.org/pdf/2404.08707v4 | 2024-06-17T11:32:29Z | 2024-04-11T17:44:56Z | Large Language Model Can Continue Evolving From Mistakes | As world knowledge evolves and new task paradigms emerge, Continual Learning (CL) is crucial for keeping Large Language Models (LLMs) up-to-date and addressing their shortcomings. In practical applications, LLMs often require both continual instruction tuning (CIT) and continual pre-training (CPT) to adapt to new task paradigms and acquire necessary knowledge for task-solving. However, it remains challenging to collect CPT data that addresses the knowledge deficiencies in models while maintaining adequate volume, and improving the efficiency of utilizing this data also presents significant difficulties. Inspired by the 'summarizing mistakes' learning skill, we propose the Continue Evolving from Mistakes (CEM) method, aiming to provide a data-efficient approach for collecting CPT data and continually improving LLMs' performance through iterative evaluation and supplementation with mistake-relevant knowledge. To efficiently utilize these CPT data and mitigate forgetting, we design a novel CL training set construction paradigm that integrates parallel CIT and CPT data. Extensive experiments demonstrate the efficacy of the CEM method, achieving up to a 17% improvement in accuracy in the best case. Furthermore, additional experiments confirm the potential of combining CEM with catastrophic forgetting mitigation methods, enabling iterative and continual model evolution. | [
"['Haokun Zhao' 'Haixia Han' 'Jie Shi' 'Chengyu Du' 'Jiaqing Liang'\n 'Yanghua Xiao']"
]
|
null | null | 2404.08708 | null | null | http://arxiv.org/pdf/2404.08708v1 | 2024-04-11T18:00:22Z | 2024-04-11T18:00:22Z | Multi-scale Topology Optimization using Neural Networks | A long-standing challenge is designing multi-scale structures with good connectivity between cells while optimizing each cell to reach close to the theoretical performance limit. We propose a new method for direct multi-scale topology optimization using neural networks. Our approach focuses on inverse homogenization that seamlessly maintains compatibility across neighboring microstructure cells. Our approach consists of a topology neural network that optimizes the microstructure shape and distribution across the design domain as a continuous field. Each microstructure cell is optimized based on a specified elasticity tensor that also accommodates in-plane rotations. The neural network takes as input the local coordinates within a cell to represent the density distribution within a cell, as well as the global coordinates of each cell to design spatially varying microstructure cells. As such, our approach models an n-dimensional multi-scale optimization problem as a 2n-dimensional inverse homogenization problem using neural networks. During the inverse homogenization of each unit cell, we extend the boundary of each cell by scaling the input coordinates such that the boundaries of neighboring cells are combined. Inverse homogenization on the combined cell improves connectivity. We demonstrate our method through the design and optimization of graded multi-scale structures. | [
"['Hongrui Chen' 'Xingchen Liu' 'Levent Burak Kara']"
]
|
null | null | 2404.08709 | null | null | http://arxiv.org/pdf/2404.08709v1 | 2024-04-11T18:07:57Z | 2024-04-11T18:07:57Z | $F_β$-plot -- a visual tool for evaluating imbalanced data
classifiers | One of the significant problems associated with imbalanced data classification is the lack of reliable metrics. This runs primarily from the fact that for most real-life (as well as commonly used benchmark) problems, we do not have information from the user on the actual form of the loss function that should be minimized. Although it is pretty common to have metrics indicating the classification quality within each class, for the end user, the analysis of several such metrics is then required, which in practice causes difficulty in interpreting the usefulness of a given classifier. Hence, many aggregate metrics have been proposed or adopted for the imbalanced data classification problem, but there is still no consensus on which should be used. An additional disadvantage is their ambiguity and systematic bias toward one class. Moreover, their use in analyzing experimental results in recognition of those classification models that perform well for the chosen aggregated metrics is burdened with the drawbacks mentioned above. Hence, the paper proposes a simple approach to analyzing the popular parametric metric $F_beta$. We point out that it is possible to indicate for a given pool of analyzed classifiers when a given model should be preferred depending on user requirements. | [
"['Szymon Wojciechowski' 'Michał Woźniak']"
]
|
null | null | 2404.08711 | null | null | http://arxiv.org/pdf/2404.08711v1 | 2024-04-11T20:11:25Z | 2024-04-11T20:11:25Z | Drug Repurposing for Parkinson's Disease Using Random Walk With Restart
Algorithm and the Parkinson's Disease Ontology Database | Parkinson's disease is a progressive and slowly developing neurodegenerative disease, characterized by dopaminergic neuron loss in the substantia nigra region of the brain. Despite extensive research by scientists, there is not yet a cure to this problem and the available therapies mainly help to reduce some of the Parkinson's symptoms. Drug repurposing (that is, the process of finding new uses for existing drugs) receives more appraisals as an efficient way that allows for reducing the time, resources, and risks associated with the development of new drugs. In this research, we design a novel computational platform that integrates gene expression data, biological networks, and the PDOD database to identify possible drug-repositioning agents for PD therapy. By using machine learning approaches like the RWR algorithm and PDOD scoring system we arrange drug-disease conversions and sort our potential sandboxes according to their possible efficacy. We propose gene expression analysis, network prioritization, and drug target data analysis to arrive at a comprehensive evaluation of drug repurposing chances. Our study results highlight such therapies as promising drug candidates to conduct further research on PD treatment. We also provide the rationale for promising drug repurposing ideas by using various sources of data and computational approaches. | [
"['Pratham Kankariya' 'Rachita Rode' 'Kevin Mudaliar'\n 'Prof. Pranali Hatode']"
]
|
null | null | 2404.08712 | null | null | http://arxiv.org/pdf/2404.08712v1 | 2024-04-11T21:04:56Z | 2024-04-11T21:04:56Z | Machine learning and economic forecasting: the role of international
trade networks | This study examines the effects of de-globalization trends on international trade networks and their role in improving forecasts for economic growth. Using section-level trade data from nearly 200 countries from 2010 to 2022, we identify significant shifts in the network topology driven by rising trade policy uncertainty. Our analysis highlights key global players through centrality rankings, with the United States, China, and Germany maintaining consistent dominance. Using a horse race of supervised regressors, we find that network topology descriptors evaluated from section-specific trade networks substantially enhance the quality of a country's GDP growth forecast. We also find that non-linear models, such as Random Forest, XGBoost, and LightGBM, outperform traditional linear models used in the economics literature. Using SHAP values to interpret these non-linear model's predictions, we find that about half of most important features originate from the network descriptors, underscoring their vital role in refining forecasts. Moreover, this study emphasizes the significance of recent economic performance, population growth, and the primary sector's influence in shaping economic growth predictions, offering novel insights into the intricacies of economic growth forecasting. | [
"['Thiago C. Silva' 'Paulo V. B. Wilhelm' 'Diego R. Amancio']"
]
|
null | null | 2404.08713 | null | null | http://arxiv.org/pdf/2404.08713v1 | 2024-04-11T21:47:13Z | 2024-04-11T21:47:13Z | Survival Prediction Across Diverse Cancer Types Using Neural Networks | Gastric cancer and Colon adenocarcinoma represent widespread and challenging malignancies with high mortality rates and complex treatment landscapes. In response to the critical need for accurate prognosis in cancer patients, the medical community has embraced the 5-year survival rate as a vital metric for estimating patient outcomes. This study introduces a pioneering approach to enhance survival prediction models for gastric and Colon adenocarcinoma patients. Leveraging advanced image analysis techniques, we sliced whole slide images (WSI) of these cancers, extracting comprehensive features to capture nuanced tumor characteristics. Subsequently, we constructed patient-level graphs, encapsulating intricate spatial relationships within tumor tissues. These graphs served as inputs for a sophisticated 4-layer graph convolutional neural network (GCN), designed to exploit the inherent connectivity of the data for comprehensive analysis and prediction. By integrating patients' total survival time and survival status, we computed C-index values for gastric cancer and Colon adenocarcinoma, yielding 0.57 and 0.64, respectively. Significantly surpassing previous convolutional neural network models, these results underscore the efficacy of our approach in accurately predicting patient survival outcomes. This research holds profound implications for both the medical and AI communities, offering insights into cancer biology and progression while advancing personalized treatment strategies. Ultimately, our study represents a significant stride in leveraging AI-driven methodologies to revolutionize cancer prognosis and improve patient outcomes on a global scale. | [
"['Xu Yan' 'Weimin Wang' 'MingXuan Xiao' 'Yufeng Li' 'Min Gao']"
]
|
null | null | 2404.08715 | null | null | http://arxiv.org/pdf/2404.08715v1 | 2024-04-12T04:14:08Z | 2024-04-12T04:14:08Z | Differentially Private Log-Location-Scale Regression Using Functional
Mechanism | This article introduces differentially private log-location-scale (DP-LLS) regression models, which incorporate differential privacy into LLS regression through the functional mechanism. The proposed models are established by injecting noise into the log-likelihood function of LLS regression for perturbed parameter estimation. We will derive the sensitivities utilized to determine the magnitude of the injected noise and prove that the proposed DP-LLS models satisfy $epsilon$-differential privacy. In addition, we will conduct simulations and case studies to evaluate the performance of the proposed models. The findings suggest that predictor dimension, training sample size, and privacy budget are three key factors impacting the performance of the proposed DP-LLS regression models. Moreover, the results indicate that a sufficiently large training dataset is needed to simultaneously ensure decent performance of the proposed models and achieve a satisfactory level of privacy protection. | [
"['Jiewen Sheng' 'Xiaolei Fang']"
]
|
null | null | 2404.08717 | null | null | http://arxiv.org/pdf/2404.08717v1 | 2024-04-12T07:32:57Z | 2024-04-12T07:32:57Z | State-Space Systems as Dynamic Generative Models | A probabilistic framework to study the dependence structure induced by deterministic discrete-time state-space systems between input and output processes is introduced. General sufficient conditions are formulated under which output processes exist and are unique once an input process has been fixed, a property that in the deterministic state-space literature is known as the echo state property. When those conditions are satisfied, the given state-space system becomes a generative model for probabilistic dependences between two sequence spaces. Moreover, those conditions guarantee that the output depends continuously on the input when using the Wasserstein metric. The output processes whose existence is proved are shown to be causal in a specific sense and to generalize those studied in purely deterministic situations. The results in this paper constitute a significant stochastic generalization of sufficient conditions for the deterministic echo state property to hold, in the sense that the stochastic echo state property can be satisfied under contractivity conditions that are strictly weaker than those in deterministic situations. This means that state-space systems can induce a purely probabilistic dependence structure between input and output sequence spaces even when there is no functional relation between those two spaces. | [
"['Juan-Pablo Ortega' 'Florian Rossmannek']"
]
|
null | null | 2404.08720 | null | null | http://arxiv.org/pdf/2404.08720v1 | 2024-04-12T11:12:16Z | 2024-04-12T11:12:16Z | Exploring Contrastive Learning for Long-Tailed Multi-Label Text
Classification | Learning an effective representation in multi-label text classification (MLTC) is a significant challenge in NLP. This challenge arises from the inherent complexity of the task, which is shaped by two key factors: the intricate connections between labels and the widespread long-tailed distribution of the data. To overcome this issue, one potential approach involves integrating supervised contrastive learning with classical supervised loss functions. Although contrastive learning has shown remarkable performance in multi-class classification, its impact in the multi-label framework has not been thoroughly investigated. In this paper, we conduct an in-depth study of supervised contrastive learning and its influence on representation in MLTC context. We emphasize the importance of considering long-tailed data distributions to build a robust representation space, which effectively addresses two critical challenges associated with contrastive learning that we identify: the "lack of positives" and the "attraction-repulsion imbalance". Building on this insight, we introduce a novel contrastive loss function for MLTC. It attains Micro-F1 scores that either match or surpass those obtained with other frequently employed loss functions, and demonstrates a significant improvement in Macro-F1 scores across three multi-label datasets. | [
"['Alexandre Audibert' 'Aurélien Gauffre' 'Massih-Reza Amini']"
]
|
null | null | 2404.08721 | null | null | http://arxiv.org/pdf/2404.08721v1 | 2024-04-12T13:11:55Z | 2024-04-12T13:11:55Z | Beyond One-Size-Fits-All: Adapting Counterfactual Explanations to User
Objectives | Explainable Artificial Intelligence (XAI) has emerged as a critical area of research aimed at enhancing the transparency and interpretability of AI systems. Counterfactual Explanations (CFEs) offer valuable insights into the decision-making processes of machine learning algorithms by exploring alternative scenarios where certain factors differ. Despite the growing popularity of CFEs in the XAI community, existing literature often overlooks the diverse needs and objectives of users across different applications and domains, leading to a lack of tailored explanations that adequately address the different use cases. In this paper, we advocate for a nuanced understanding of CFEs, recognizing the variability in desired properties based on user objectives and target applications. We identify three primary user objectives and explore the desired characteristics of CFEs in each case. By addressing these differences, we aim to design more effective and tailored explanations that meet the specific needs of users, thereby enhancing collaboration with AI systems. | [
"['Orfeas Menis Mastromichalakis' 'Jason Liartis' 'Giorgos Stamou']"
]
|
null | null | 2404.08722 | null | null | http://arxiv.org/pdf/2404.08722v2 | 2024-06-26T16:46:19Z | 2024-04-12T13:24:28Z | VADA: a Data-Driven Simulator for Nanopore Sequencing | Nanopore sequencing offers the ability for real-time analysis of long DNA sequences at a low cost, enabling new applications such as early detection of cancer. Due to the complex nature of nanopore measurements and the high cost of obtaining ground truth datasets, there is a need for nanopore simulators. Existing simulators rely on handcrafted rules and parameters and do not learn an internal representation that would allow for analysing underlying biological factors of interest. Instead, we propose VADA, a purely data-driven method for simulating nanopores based on an autoregressive latent variable model. We embed subsequences of DNA and introduce a conditional prior to address the challenge of a collapsing conditioning. We introduce an auxiliary regressor on the latent variable to encourage our model to learn an informative latent representation. We empirically demonstrate that our model achieves competitive simulation performance on experimental nanopore data. Moreover, we show we have learned an informative latent representation that is predictive of the DNA labels. We hypothesize that other biological factors of interest, beyond the DNA labels, can potentially be extracted from such a learned latent representation. | [
"['Jonas Niederle' 'Simon Koop' 'Marc Pagès-Gallego' 'Vlado Menkovski']"
]
|
null | null | 2404.08747 | null | null | http://arxiv.org/pdf/2404.08747v1 | 2024-04-12T18:20:26Z | 2024-04-12T18:20:26Z | Observation-specific explanations through scattered data approximation | This work introduces the definition of observation-specific explanations to assign a score to each data point proportional to its importance in the definition of the prediction process. Such explanations involve the identification of the most influential observations for the black-box model of interest. The proposed method involves estimating these explanations by constructing a surrogate model through scattered data approximation utilizing the orthogonal matching pursuit algorithm. The proposed approach is validated on both simulated and real-world datasets. | [
"['Valentina Ghidini' 'Michael Multerer' 'Jacopo Quizi' 'Rohan Sen']"
]
|
null | null | 2404.08750 | null | null | http://arxiv.org/pdf/2404.08750v1 | 2024-04-12T18:23:29Z | 2024-04-12T18:23:29Z | FastLogAD: Log Anomaly Detection with Mask-Guided Pseudo Anomaly
Generation and Discrimination | Nowadays large computers extensively output logs to record the runtime status and it has become crucial to identify any suspicious or malicious activities from the information provided by the realtime logs. Thus, fast log anomaly detection is a necessary task to be implemented for automating the infeasible manual detection. Most of the existing unsupervised methods are trained only on normal log data, but they usually require either additional abnormal data for hyperparameter selection or auxiliary datasets for discriminative model optimization. In this paper, aiming for a highly effective discriminative model that enables rapid anomaly detection,we propose FastLogAD, a generator-discriminator framework trained to exhibit the capability of generating pseudo-abnormal logs through the Mask-Guided Anomaly Generation (MGAG) model and efficiently identifying the anomalous logs via the Discriminative Abnormality Separation (DAS) model. Particularly, pseudo-abnormal logs are generated by replacing randomly masked tokens in a normal sequence with unlikely candidates. During the discriminative stage, FastLogAD learns a distinct separation between normal and pseudoabnormal samples based on their embedding norms, allowing the selection of a threshold without exposure to any test data and achieving competitive performance. Extensive experiments on several common benchmarks show that our proposed FastLogAD outperforms existing anomaly detection approaches. Furthermore, compared to previous methods, FastLogAD achieves at least x10 speed increase in anomaly detection over prior work. Our implementation is available at https://github.com/YifeiLin0226/FastLogAD. | [
"['Yifei Lin' 'Hanqiu Deng' 'Xingyu Li']"
]
|
null | null | 2404.08754 | null | null | http://arxiv.org/pdf/2404.08754v1 | 2024-04-12T18:26:32Z | 2024-04-12T18:26:32Z | Computing distances and means on manifolds with a metric-constrained
Eikonal approach | Computing distances on Riemannian manifolds is a challenging problem with numerous applications, from physics, through statistics, to machine learning. In this paper, we introduce the metric-constrained Eikonal solver to obtain continuous, differentiable representations of distance functions on manifolds. The differentiable nature of these representations allows for the direct computation of globally length-minimising paths on the manifold. We showcase the use of metric-constrained Eikonal solvers for a range of manifolds and demonstrate the applications. First, we demonstrate that metric-constrained Eikonal solvers can be used to obtain the Fr'echet mean on a manifold, employing the definition of a Gaussian mixture model, which has an analytical solution to verify the numerical results. Second, we demonstrate how the obtained distance function can be used to conduct unsupervised clustering on the manifold -- a task for which existing approaches are computationally prohibitive. This work opens opportunities for distance computations on manifolds. | [
"['Daniel Kelshaw' 'Luca Magri']"
]
|
null | null | 2404.08755 | null | null | http://arxiv.org/pdf/2404.08755v1 | 2024-04-12T18:28:44Z | 2024-04-12T18:28:44Z | Training a Vision Language Model as Smartphone Assistant | Addressing the challenge of a digital assistant capable of executing a wide array of user tasks, our research focuses on the realm of instruction-based mobile device control. We leverage recent advancements in large language models (LLMs) and present a visual language model (VLM) that can fulfill diverse tasks on mobile devices. Our model functions by interacting solely with the user interface (UI). It uses the visual input from the device screen and mimics human-like interactions, encompassing gestures such as tapping and swiping. This generality in the input and output space allows our agent to interact with any application on the device. Unlike previous methods, our model operates not only on a single screen image but on vision-language sentences created from sequences of past screenshots along with corresponding actions. Evaluating our method on the challenging Android in the Wild benchmark demonstrates its promising efficacy and potential. | [
"['Nicolai Dorka' 'Janusz Marecki' 'Ammar Anwar']"
]
|
null | null | 2404.08761 | null | null | http://arxiv.org/pdf/2404.08761v1 | 2024-04-12T18:37:00Z | 2024-04-12T18:37:00Z | `Eyes of a Hawk and Ears of a Fox': Part Prototype Network for
Generalized Zero-Shot Learning | Current approaches in Generalized Zero-Shot Learning (GZSL) are built upon base models which consider only a single class attribute vector representation over the entire image. This is an oversimplification of the process of novel category recognition, where different regions of the image may have properties from different seen classes and thus have different predominant attributes. With this in mind, we take a fundamentally different approach: a pre-trained Vision-Language detector (VINVL) sensitive to attribute information is employed to efficiently obtain region features. A learned function maps the region features to region-specific attribute attention used to construct class part prototypes. We conduct experiments on a popular GZSL benchmark consisting of the CUB, SUN, and AWA2 datasets where our proposed Part Prototype Network (PPN) achieves promising results when compared with other popular base models. Corresponding ablation studies and analysis show that our approach is highly practical and has a distinct advantage over global attribute attention when localized proposals are available. | [
"['Joshua Feinglass' 'Jayaraman J. Thiagarajan' 'Rushil Anirudh'\n 'T. S. Jayram' 'Yezhou Yang']"
]
|
null | null | 2404.08763 | null | null | http://arxiv.org/pdf/2404.08763v2 | 2024-04-27T00:01:02Z | 2024-04-12T18:42:18Z | CATS: Contextually-Aware Thresholding for Sparsity in Large Language
Models | Large Language Models (LLMs) have dramatically advanced AI applications, yet their deployment remains challenging due to their immense inference costs. Recent studies ameliorate the computational costs of LLMs by increasing their activation sparsity but suffer from significant performance degradation on downstream tasks. In this work, we introduce a new framework for sparsifying the activations of base LLMs and reducing inference costs, dubbed Contextually Aware Thresholding for Sparsity (CATS). CATS is relatively simple, easy to implement, and highly effective. At the heart of our framework is a new non-linear activation function. We demonstrate that CATS can be applied to various base models, including Mistral-7B and Llama2-7B, and outperforms existing sparsification techniques in downstream task performance. More precisely, CATS-based models often achieve downstream task performance within 1-2% of their base models without any fine-tuning and even at activation sparsity levels of 50%. Furthermore, CATS-based models converge faster and display better task performance than competing techniques when fine-tuning is applied. Finally, we develop a custom GPU kernel for efficient implementation of CATS that translates the activation of sparsity of CATS to real wall-clock time speedups. Our custom kernel implementation of CATS results in a ~15% improvement in wall-clock inference latency of token generation on both Llama-7B and Mistral-7B. | [
"['Je-Yong Lee' 'Donghyun Lee' 'Genghan Zhang' 'Mo Tiwari'\n 'Azalia Mirhoseini']"
]
|
null | null | 2404.08767 | null | null | http://arxiv.org/pdf/2404.08767v1 | 2024-04-12T18:45:51Z | 2024-04-12T18:45:51Z | LLM-Seg: Bridging Image Segmentation and Large Language Model Reasoning | Understanding human instructions to identify the target objects is vital for perception systems. In recent years, the advancements of Large Language Models (LLMs) have introduced new possibilities for image segmentation. In this work, we delve into reasoning segmentation, a novel task that enables segmentation system to reason and interpret implicit user intention via large language model reasoning and then segment the corresponding target. Our work on reasoning segmentation contributes on both the methodological design and dataset labeling. For the model, we propose a new framework named LLM-Seg. LLM-Seg effectively connects the current foundational Segmentation Anything Model and the LLM by mask proposals selection. For the dataset, we propose an automatic data generation pipeline and construct a new reasoning segmentation dataset named LLM-Seg40K. Experiments demonstrate that our LLM-Seg exhibits competitive performance compared with existing methods. Furthermore, our proposed pipeline can efficiently produce high-quality reasoning segmentation datasets. The LLM-Seg40K dataset, developed through this pipeline, serves as a new benchmark for training and evaluating various reasoning segmentation approaches. Our code, models and dataset are at https://github.com/wangjunchi/LLMSeg. | [
"['Junchi Wang' 'Lei Ke']"
]
|
null | null | 2404.08778 | null | null | http://arxiv.org/abs/2404.08778v1 | 2024-04-12T19:04:59Z | 2024-04-12T19:04:59Z | Towards Sim-to-Real Industrial Parts Classification with Synthetic
Dataset | This paper is about effectively utilizing synthetic data for training deep neural networks for industrial parts classification, in particular, by taking into account the domain gap against real-world images. To this end, we introduce a synthetic dataset that may serve as a preliminary testbed for the Sim-to-Real challenge; it contains 17 objects of six industrial use cases, including isolated and assembled parts. A few subsets of objects exhibit large similarities in shape and albedo for reflecting challenging cases of industrial parts. All the sample images come with and without random backgrounds and post-processing for evaluating the importance of domain randomization. We call it Synthetic Industrial Parts dataset (SIP-17). We study the usefulness of SIP-17 through benchmarking the performance of five state-of-the-art deep network models, supervised and self-supervised, trained only on the synthetic data while testing them on real data. By analyzing the results, we deduce some insights on the feasibility and challenges of using synthetic data for industrial parts classification and for further developing larger-scale synthetic datasets. Our dataset and code are publicly available. | [
"['Xiaomeng Zhu' 'Talha Bilal' 'Pär Mårtensson' 'Lars Hanson'\n 'Mårten Björkman' 'Atsuto Maki']"
]
|
null | null | 2404.08785 | null | null | http://arxiv.org/pdf/2404.08785v1 | 2024-04-12T19:13:42Z | 2024-04-12T19:13:42Z | Under pressure: learning-based analog gauge reading in the wild | We propose an interpretable framework for reading analog gauges that is deployable on real world robotic systems. Our framework splits the reading task into distinct steps, such that we can detect potential failures at each step. Our system needs no prior knowledge of the type of gauge or the range of the scale and is able to extract the units used. We show that our gauge reading algorithm is able to extract readings with a relative reading error of less than 2%. | [
"['Maurits Reitsma' 'Julian Keller' 'Kenneth Blomqvist' 'Roland Siegwart']"
]
|
null | null | 2404.08788 | null | null | http://arxiv.org/pdf/2404.08788v1 | 2024-04-12T19:29:10Z | 2024-04-12T19:29:10Z | Detecting AI-Generated Images via CLIP | As AI-generated image (AIGI) methods become more powerful and accessible, it has become a critical task to determine if an image is real or AI-generated. Because AIGI lack the signatures of photographs and have their own unique patterns, new models are needed to determine if an image is AI-generated. In this paper, we investigate the ability of the Contrastive Language-Image Pre-training (CLIP) architecture, pre-trained on massive internet-scale data sets, to perform this differentiation. We fine-tune CLIP on real images and AIGI from several generative models, enabling CLIP to determine if an image is AI-generated and, if so, determine what generation method was used to create it. We show that the fine-tuned CLIP architecture is able to differentiate AIGI as well or better than models whose architecture is specifically designed to detect AIGI. Our method will significantly increase access to AIGI-detecting tools and reduce the negative effects of AIGI on society, as our CLIP fine-tuning procedures require no architecture changes from publicly available model repositories and consume significantly less GPU resources than other AIGI detection models. | [
"['A. G. Moskowitz' 'T. Gaona' 'J. Peterson']"
]
|
null | null | 2404.08789 | null | null | http://arxiv.org/pdf/2404.08789v1 | 2024-04-12T19:33:52Z | 2024-04-12T19:33:52Z | Differentiable and Stable Long-Range Tracking of Multiple Posterior
Modes | Particle filters flexibly represent multiple posterior modes nonparametrically, via a collection of weighted samples, but have classically been applied to tracking problems with known dynamics and observation likelihoods. Such generative models may be inaccurate or unavailable for high-dimensional observations like images. We instead leverage training data to discriminatively learn particle-based representations of uncertainty in latent object states, conditioned on arbitrary observations via deep neural network encoders. While prior discriminative particle filters have used heuristic relaxations of discrete particle resampling, or biased learning by truncating gradients at resampling steps, we achieve unbiased and low-variance gradient estimates by representing posteriors as continuous mixture densities. Our theory and experiments expose dramatic failures of existing reparameterization-based estimators for mixture gradients, an issue we address via an importance-sampling gradient estimator. Unlike standard recurrent neural networks, our mixture density particle filter represents multimodal uncertainty in continuous latent states, improving accuracy and robustness. On a range of challenging tracking and robot localization problems, our approach achieves dramatic improvements in accuracy, while also showing much greater stability across multiple training runs. | [
"['Ali Younis' 'Erik Sudderth']"
]
|
null | null | 2404.08791 | null | null | http://arxiv.org/pdf/2404.08791v1 | 2024-04-12T19:43:37Z | 2024-04-12T19:43:37Z | Handling Reward Misspecification in the Presence of Expectation Mismatch | Detecting and handling misspecified objectives, such as reward functions, has been widely recognized as one of the central challenges within the domain of Artificial Intelligence (AI) safety research. However, even with the recognition of the importance of this problem, we are unaware of any works that attempt to provide a clear definition for what constitutes (a) misspecified objectives and (b) successfully resolving such misspecifications. In this work, we use the theory of mind, i.e., the human user's beliefs about the AI agent, as a basis to develop a formal explanatory framework called Expectation Alignment (EAL) to understand the objective misspecification and its causes. Our EAL framework not only acts as an explanatory framework for existing works but also provides us with concrete insights into the limitations of existing methods to handle reward misspecification and novel solution strategies. We use these insights to propose a new interactive algorithm that uses the specified reward to infer potential user expectations about the system behavior. We show how one can efficiently implement this algorithm by mapping the inference problem into linear programs. We evaluate our method on a set of standard Markov Decision Process (MDP) benchmarks. | [
"['Sarath Sreedharan' 'Malek Mechergui']"
]
|
null | null | 2404.08792 | null | null | http://arxiv.org/pdf/2404.08792v1 | 2024-04-12T19:43:54Z | 2024-04-12T19:43:54Z | Convergence of coordinate ascent variational inference for log-concave
measures via optimal transport | Mean field variational inference (VI) is the problem of finding the closest product (factorized) measure, in the sense of relative entropy, to a given high-dimensional probability measure $rho$. The well known Coordinate Ascent Variational Inference (CAVI) algorithm aims to approximate this product measure by iteratively optimizing over one coordinate (factor) at a time, which can be done explicitly. Despite its popularity, the convergence of CAVI remains poorly understood. In this paper, we prove the convergence of CAVI for log-concave densities $rho$. If additionally $log rho$ has Lipschitz gradient, we find a linear rate of convergence, and if also $rho$ is strongly log-concave, we find an exponential rate. Our analysis starts from the observation that mean field VI, while notoriously non-convex in the usual sense, is in fact displacement convex in the sense of optimal transport when $rho$ is log-concave. This allows us to adapt techniques from the optimization literature on coordinate descent algorithms in Euclidean space. | [
"['Manuel Arnese' 'Daniel Lacker']"
]
|
null | null | 2404.08797 | null | null | http://arxiv.org/pdf/2404.08797v1 | 2024-04-12T20:13:19Z | 2024-04-12T20:13:19Z | Diffusion-Based Joint Temperature and Precipitation Emulation of Earth
System Models | Earth system models (ESMs) are the principal tools used in climate science to generate future climate projections under various atmospheric emissions scenarios on a global or regional scale. Generative deep learning approaches are suitable for emulating these tools due to their computational efficiency and ability, once trained, to generate realizations in a fraction of the time required by ESMs. We extend previous work that used a generative probabilistic diffusion model to emulate ESMs by targeting the joint emulation of multiple variables, temperature and precipitation, by a single diffusion model. Joint generation of multiple variables is critical to generate realistic samples of phenomena resulting from the interplay of multiple variables. The diffusion model emulator takes in the monthly mean-maps of temperature and precipitation and produces the daily values of each of these variables that exhibit statistical properties similar to those generated by ESMs. Our results show the outputs from our extended model closely resemble those from ESMs on various climate metrics including dry spells and hot streaks, and that the joint distribution of temperature and precipitation in our sample closely matches those of ESMs. | [
"['Katie Christensen' 'Lyric Otto' 'Seth Bassetti' 'Claudia Tebaldi'\n 'Brian Hutchinson']"
]
|
null | null | 2404.08799 | null | null | http://arxiv.org/pdf/2404.08799v1 | 2024-04-12T20:16:03Z | 2024-04-12T20:16:03Z | Semantic Approach to Quantifying the Consistency of Diffusion Model
Image Generation | In this study, we identify the need for an interpretable, quantitative score of the repeatability, or consistency, of image generation in diffusion models. We propose a semantic approach, using a pairwise mean CLIP (Contrastive Language-Image Pretraining) score as our semantic consistency score. We applied this metric to compare two state-of-the-art open-source image generation diffusion models, Stable Diffusion XL and PixArt-{alpha}, and we found statistically significant differences between the semantic consistency scores for the models. Agreement between the Semantic Consistency Score selected model and aggregated human annotations was 94%. We also explored the consistency of SDXL and a LoRA-fine-tuned version of SDXL and found that the fine-tuned model had significantly higher semantic consistency in generated images. The Semantic Consistency Score proposed here offers a measure of image generation alignment, facilitating the evaluation of model architectures for specific tasks and aiding in informed decision-making regarding model selection. | [
"['Brinnae Bent']"
]
|
null | null | 2404.08801 | null | null | http://arxiv.org/pdf/2404.08801v2 | 2024-04-16T07:27:58Z | 2024-04-12T20:28:14Z | Megalodon: Efficient LLM Pretraining and Inference with Unlimited
Context Length | The quadratic complexity and weak length extrapolation of Transformers limits their ability to scale to long sequences, and while sub-quadratic solutions like linear attention and state space models exist, they empirically underperform Transformers in pretraining efficiency and downstream task accuracy. We introduce Megalodon, a neural architecture for efficient sequence modeling with unlimited context length. Megalodon inherits the architecture of Mega (exponential moving average with gated attention), and further introduces multiple technical components to improve its capability and stability, including complex exponential moving average (CEMA), timestep normalization layer, normalized attention mechanism and pre-norm with two-hop residual configuration. In a controlled head-to-head comparison with Llama2, Megalodon achieves better efficiency than Transformer in the scale of 7 billion parameters and 2 trillion training tokens. Megalodon reaches a training loss of 1.70, landing mid-way between Llama2-7B (1.75) and 13B (1.67). Code: https://github.com/XuezheMax/megalodon | [
"['Xuezhe Ma' 'Xiaomeng Yang' 'Wenhan Xiong' 'Beidi Chen' 'Lili Yu'\n 'Hao Zhang' 'Jonathan May' 'Luke Zettlemoyer' 'Omer Levy' 'Chunting Zhou']"
]
|
null | null | 2404.08805 | null | null | http://arxiv.org/abs/2404.08805v1 | 2024-04-12T20:39:19Z | 2024-04-12T20:39:19Z | Real-time guidewire tracking and segmentation in intraoperative x-ray | During endovascular interventions, physicians have to perform accurate and immediate operations based on the available real-time information, such as the shape and position of guidewires observed on the fluoroscopic images, haptic information and the patients' physiological signals. For this purpose, real-time and accurate guidewire segmentation and tracking can enhance the visualization of guidewires and provide visual feedback for physicians during the intervention as well as for robot-assisted interventions. Nevertheless, this task often comes with the challenge of elongated deformable structures that present themselves with low contrast in the noisy fluoroscopic image sequences. To address these issues, a two-stage deep learning framework for real-time guidewire segmentation and tracking is proposed. In the first stage, a Yolov5s detector is trained, using the original X-ray images as well as synthetic ones, which is employed to output the bounding boxes of possible target guidewires. More importantly, a refinement module based on spatiotemporal constraints is incorporated to robustly localize the guidewire and remove false detections. In the second stage, a novel and efficient network is proposed to segment the guidewire in each detected bounding box. The network contains two major modules, namely a hessian-based enhancement embedding module and a dual self-attention module. Quantitative and qualitative evaluations on clinical intra-operative images demonstrate that the proposed approach significantly outperforms our baselines as well as the current state of the art and, in comparison, shows higher robustness to low quality images. | [
"['Baochang Zhang' 'Mai Bui' 'Cheng Wang' 'Felix Bourier'\n 'Heribert Schunkert' 'Nassir Navab']"
]
|
null | null | 2404.08809 | null | null | http://arxiv.org/pdf/2404.08809v1 | 2024-04-12T20:54:01Z | 2024-04-12T20:54:01Z | Leveraging viscous Hamilton-Jacobi PDEs for uncertainty quantification
in scientific machine learning | Uncertainty quantification (UQ) in scientific machine learning (SciML) combines the powerful predictive power of SciML with methods for quantifying the reliability of the learned models. However, two major challenges remain: limited interpretability and expensive training procedures. We provide a new interpretation for UQ problems by establishing a new theoretical connection between some Bayesian inference problems arising in SciML and viscous Hamilton-Jacobi partial differential equations (HJ PDEs). Namely, we show that the posterior mean and covariance can be recovered from the spatial gradient and Hessian of the solution to a viscous HJ PDE. As a first exploration of this connection, we specialize to Bayesian inference problems with linear models, Gaussian likelihoods, and Gaussian priors. In this case, the associated viscous HJ PDEs can be solved using Riccati ODEs, and we develop a new Riccati-based methodology that provides computational advantages when continuously updating the model predictions. Specifically, our Riccati-based approach can efficiently add or remove data points to the training set invariant to the order of the data and continuously tune hyperparameters. Moreover, neither update requires retraining on or access to previously incorporated data. We provide several examples from SciML involving noisy data and textit{epistemic uncertainty} to illustrate the potential advantages of our approach. In particular, this approach's amenability to data streaming applications demonstrates its potential for real-time inferences, which, in turn, allows for applications in which the predicted uncertainty is used to dynamically alter the learning process. | [
"['Zongren Zou' 'Tingwei Meng' 'Paula Chen' 'Jérôme Darbon'\n 'George Em Karniadakis']"
]
|
null | null | 2404.08811 | null | null | http://arxiv.org/pdf/2404.08811v1 | 2024-04-12T20:58:25Z | 2024-04-12T20:58:25Z | Reducing the Barriers to Entry for Foundation Model Training | The world has recently witnessed an unprecedented acceleration in demands for Machine Learning and Artificial Intelligence applications. This spike in demand has imposed tremendous strain on the underlying technology stack in supply chain, GPU-accelerated hardware, software, datacenter power density, and energy consumption. If left on the current technological trajectory, future demands show insurmountable spending trends, further limiting market players, stifling innovation, and widening the technology gap. To address these challenges, we propose a fundamental change in the AI training infrastructure throughout the technology ecosystem. The changes require advancements in supercomputing and novel AI training approaches, from high-end software to low-level hardware, microprocessor, and chip design, while advancing the energy efficiency required by a sustainable infrastructure. This paper presents the analytical framework that quantitatively highlights the challenges and points to the opportunities to reduce the barriers to entry for training large language models. | [
"['Paolo Faraboschi' 'Ellis Giles' 'Justin Hotard' 'Konstanty Owczarek'\n 'Andrew Wheeler']"
]
|
null | null | 2404.08814 | null | null | http://arxiv.org/pdf/2404.08814v2 | 2024-04-16T14:17:51Z | 2024-04-12T21:14:20Z | E3: Ensemble of Expert Embedders for Adapting Synthetic Image Detectors
to New Generators Using Limited Data | As generative AI progresses rapidly, new synthetic image generators continue to emerge at a swift pace. Traditional detection methods face two main challenges in adapting to these generators: the forensic traces of synthetic images from new techniques can vastly differ from those learned during training, and access to data for these new generators is often limited. To address these issues, we introduce the Ensemble of Expert Embedders (E3), a novel continual learning framework for updating synthetic image detectors. E3 enables the accurate detection of images from newly emerged generators using minimal training data. Our approach does this by first employing transfer learning to develop a suite of expert embedders, each specializing in the forensic traces of a specific generator. Then, all embeddings are jointly analyzed by an Expert Knowledge Fusion Network to produce accurate and reliable detection decisions. Our experiments demonstrate that E3 outperforms existing continual learning methods, including those developed specifically for synthetic image detection. | [
"['Aref Azizpour' 'Tai D. Nguyen' 'Manil Shrestha' 'Kaidi Xu' 'Edward Kim'\n 'Matthew C. Stamm']"
]
|
null | null | 2404.08819 | null | null | http://arxiv.org/pdf/2404.08819v2 | 2024-06-04T22:05:45Z | 2024-04-12T21:30:06Z | The Illusion of State in State-Space Models | State-space models (SSMs) have emerged as a potential alternative architecture for building large language models (LLMs) compared to the previously ubiquitous transformer architecture. One theoretical weakness of transformers is that they cannot express certain kinds of sequential computation and state tracking (Merrill & Sabharwal, 2023), which SSMs are explicitly designed to address via their close architectural similarity to recurrent neural networks (RNNs). But do SSMs truly have an advantage (over transformers) in expressive power for state tracking? Surprisingly, the answer is no. Our analysis reveals that the expressive power of SSMs is limited very similarly to transformers: SSMs cannot express computation outside the complexity class $mathsf{TC}^0$. In particular, this means they cannot solve simple state-tracking problems like permutation composition. It follows that SSMs are provably unable to accurately track chess moves with certain notation, evaluate code, or track entities in a long narrative. To supplement our formal analysis, we report experiments showing that Mamba-style SSMs indeed struggle with state tracking. Thus, despite its recurrent formulation, the "state" in an SSM is an illusion: SSMs have similar expressiveness limitations to non-recurrent models like transformers, which may fundamentally limit their ability to solve real-world state-tracking problems. | [
"['William Merrill' 'Jackson Petty' 'Ashish Sabharwal']"
]
|
null | null | 2404.08820 | null | null | http://arxiv.org/pdf/2404.08820v1 | 2024-04-12T21:30:09Z | 2024-04-12T21:30:09Z | Single-image driven 3d viewpoint training data augmentation for
effective wine label recognition | Confronting the critical challenge of insufficient training data in the field of complex image recognition, this paper introduces a novel 3D viewpoint augmentation technique specifically tailored for wine label recognition. This method enhances deep learning model performance by generating visually realistic training samples from a single real-world wine label image, overcoming the challenges posed by the intricate combinations of text and logos. Classical Generative Adversarial Network (GAN) methods fall short in synthesizing such intricate content combination. Our proposed solution leverages time-tested computer vision and image processing strategies to expand our training dataset, thereby broadening the range of training samples for deep learning applications. This innovative approach to data augmentation circumvents the constraints of limited training resources. Using the augmented training images through batch-all triplet metric learning on a Vision Transformer (ViT) architecture, we can get the most discriminative embedding features for every wine label, enabling us to perform one-shot recognition of existing wine labels in the training classes or future newly collected wine labels unavailable in the training. Experimental results show a significant increase in recognition accuracy over conventional 2D data augmentation techniques. | [
"['Yueh-Cheng Huang' 'Hsin-Yi Chen' 'Cheng-Jui Hung' 'Jen-Hui Chuang'\n 'Jenq-Neng Hwang']"
]
|
null | null | 2404.08828 | null | null | http://arxiv.org/pdf/2404.08828v1 | 2024-04-12T21:59:42Z | 2024-04-12T21:59:42Z | Hindsight PRIORs for Reward Learning from Human Preferences | Preference based Reinforcement Learning (PbRL) removes the need to hand specify a reward function by learning a reward from preference feedback over policy behaviors. Current approaches to PbRL do not address the credit assignment problem inherent in determining which parts of a behavior most contributed to a preference, which result in data intensive approaches and subpar reward functions. We address such limitations by introducing a credit assignment strategy (Hindsight PRIOR) that uses a world model to approximate state importance within a trajectory and then guides rewards to be proportional to state importance through an auxiliary predicted return redistribution objective. Incorporating state importance into reward learning improves the speed of policy learning, overall policy performance, and reward recovery on both locomotion and manipulation tasks. For example, Hindsight PRIOR recovers on average significantly (p<0.05) more reward on MetaWorld (20%) and DMC (15%). The performance gains and our ablations demonstrate the benefits even a simple credit assignment strategy can have on reward learning and that state importance in forward dynamics prediction is a strong proxy for a state's contribution to a preference decision. Code repository can be found at https://github.com/apple/ml-rlhf-hindsight-prior. | [
"['Mudit Verma' 'Katherine Metcalf']"
]
|
null | null | 2404.08829 | null | null | http://arxiv.org/pdf/2404.08829v1 | 2024-04-12T22:00:27Z | 2024-04-12T22:00:27Z | Measuring the Predictability of Recommender Systems using Structural
Complexity Metrics | Recommender systems (RS) are central to the filtering and curation of online content. These algorithms predict user ratings for unseen items based on past preferences. Despite their importance, the innate predictability of RS has received limited attention. This study introduces data-driven metrics to measure the predictability of RS based on the structural complexity of the user-item rating matrix. A low predictability score indicates complex and unpredictable user-item interactions, while a high predictability score reveals less complex patterns with predictive potential. We propose two strategies that use singular value decomposition (SVD) and matrix factorization (MF) to measure structural complexity. By perturbing the data and evaluating the prediction of the perturbed version, we explore the structural consistency indicated by the SVD singular vectors. The assumption is that a random perturbation of highly structured data does not change its structure. Empirical results show a high correlation between our metrics and the accuracy of the best-performing prediction algorithms on real data sets. | [
"['Alfonso Valderrama' 'Andrés Abeliuk']"
]
|
null | null | 2404.08831 | null | null | http://arxiv.org/pdf/2404.08831v1 | 2024-04-12T22:05:01Z | 2024-04-12T22:05:01Z | Structured Model Pruning for Efficient Inference in Computational
Pathology | Recent years have seen significant efforts to adopt Artificial Intelligence (AI) in healthcare for various use cases, from computer-aided diagnosis to ICU triage. However, the size of AI models has been rapidly growing due to scaling laws and the success of foundational models, which poses an increasing challenge to leverage advanced models in practical applications. It is thus imperative to develop efficient models, especially for deploying AI solutions under resource-constrains or with time sensitivity. One potential solution is to perform model compression, a set of techniques that remove less important model components or reduce parameter precision, to reduce model computation demand. In this work, we demonstrate that model pruning, as a model compression technique, can effectively reduce inference cost for computational and digital pathology based analysis with a negligible loss of analysis performance. To this end, we develop a methodology for pruning the widely used U-Net-style architectures in biomedical imaging, with which we evaluate multiple pruning heuristics on nuclei instance segmentation and classification, and empirically demonstrate that pruning can compress models by at least 70% with a negligible drop in performance. | [
"['Mohammed Adnan' 'Qinle Ba' 'Nazim Shaikh' 'Shivam Kalra'\n 'Satarupa Mukherjee' 'Auranuch Lorsakul']"
]
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.