English
Related papers

Related papers: Grokking vs. Learning: Same Features, Different En…

200 papers

Systematic Generalization refers to a learning algorithm's ability to extrapolate learned behavior to unseen situations that are distinct but semantically similar to its training data. As shown in recent work, state-of-the-art deep learning…

Artificial Intelligence · Computer Science 2020-10-06 Tong Gao , Qi Huang , Raymond J. Mooney

The fitness landscape metaphor plays a central role on the modeling of optimizing principles in many research fields, ranging from evolutionary biology, where it was first introduced, to management research. Here we consider the ensemble of…

Populations and Evolution · Quantitative Biology 2019-01-30 Paulo R. A. Campos , José F. Fontanari

Recent advances in physics-augmented neural networks have enabled thermodynamically consistent data-driven constitutive modeling of complex inelastic materials. Most existing approaches, however, implicitly adopt a specific thermodynamic…

Materials Science · Physics 2026-05-28 Reese E. Jones , Jan N. Fuhg

Recent work has shown that learned image compression strategies can outperform standard hand-crafted compression algorithms that have been developed over decades of intensive research on the rate-distortion trade-off. With growing…

Image and Video Processing · Electrical Eng. & Systems 2021-11-04 Felipe Codevilla , Jean Gabriel Simard , Ross Goroshin , Chris Pal

Recent developments in sequential experimental design look to construct a policy that can efficiently navigate the design space, in a way that maximises the expected information gain. Whilst there is work on achieving tractable policies for…

Machine Learning · Computer Science 2025-08-20 Yasir Zubayr Barlas , Kizito Salako

The study of Deep Network (DN) training dynamics has largely focused on the evolution of the loss function, evaluated on or around train and test set data points. In fact, many DN phenomenon were first introduced in literature with that…

Machine Learning · Computer Science 2023-10-23 Ahmed Imtiaz Humayun , Randall Balestriero , Richard Baraniuk

Learning to locomote to arbitrary goals on hardware remains a challenging problem for reinforcement learning. In this paper, we present a hierarchical learning framework that improves sample-efficiency and generalizability of locomotion…

Robotics · Computer Science 2019-09-30 Tianyu Li , Nathan Lambert , Roberto Calandra , Franziska Meier , Akshara Rai

Diffusion models (DMs) are a powerful generative framework that have attracted significant attention in recent years. However, the high computational cost of training DMs limits their practical applications. In this paper, we start with a…

Machine Learning · Computer Science 2024-04-12 Tianshuo Xu , Peng Mi , Ruilin Wang , Yingcong Chen

Grokking - the delayed transition from memorisation to generalisation in neural networks - remains poorly understood. We study this phenomenon through the geometry of learned representations and identify a consistent empirical signature…

Machine Learning · Computer Science 2026-05-13 Truong Xuan Khanh , Truong Quynh Hoa , Luu Duc Trung , Phan Thanh Duc

This thesis explores how deep learning models learn over time, using ideas inspired by force analysis. Specifically, we zoom in on the model's training procedure to see how one training example affects another during learning, like…

Machine Learning · Computer Science 2025-09-25 Yi Ren

Neural networks trained to solve modular arithmetic tasks exhibit grokking, a phenomenon where the test accuracy starts improving long after the model achieves 100% training accuracy in the training process. It is often taken as an example…

This study investigates the impact of gradient compression on distributed training performance, focusing on sparsification and quantization techniques, including top-k, DGC, and QSGD. In baseline experiments, random-k compression results in…

Machine Learning · Computer Science 2025-02-12 Shruti Singh , Shantanu Kumar

Curriculum learning allows complex tasks to be mastered via incremental progression over `stepping stone' goals towards a final desired behaviour. Typical implementations learn locomotion policies for challenging environments through…

Neural and Evolutionary Computing · Computer Science 2022-03-30 David Howard , Josh Kannemeyer , Davide Dolcetti , Humphrey Munn , Nicole Robinson

In many scientific and data-driven applications, machine learning models are increasingly used as measurement instruments, rather than merely as predictors of predefined labels. When the measurement function is learned from data, the…

Machine Learning · Computer Science 2026-01-27 Indrė Žliobaitė

This work focuses on the training dynamics of one associative memory module storing outer products of token embeddings. We reduce this problem to the study of a system of particles, which interact according to properties of the data…

Machine Learning · Computer Science 2024-03-01 Vivien Cabannes , Berfin Simsek , Alberto Bietti

In the Mixup training paradigm, a model is trained using convex combinations of data points and their associated labels. Despite seeing very few true data points during training, models trained using Mixup seem to still minimize the…

Machine Learning · Computer Science 2022-02-22 Muthu Chidambaram , Xiang Wang , Yuzheng Hu , Chenwei Wu , Rong Ge

Recent studies on transformer-based language models show that they can answer questions by reasoning over knowledge provided as part of the context (i.e., in-context reasoning). However, since the available knowledge is often not filtered…

Computation and Language · Computer Science 2023-11-07 Zeming Chen , Gail Weiss , Eric Mitchell , Asli Celikyilmaz , Antoine Bosselut

Treeging combines the flexible mean structure of regression trees with the covariance-based prediction strategy of kriging into the base learner of an ensemble prediction algorithm. In so doing, it combines the strengths of the two primary…

Machine Learning · Statistics 2021-10-05 Gregory L. Watson , Michael Jerrett , Colleen E. Reid , Donatello Telesca

This article considers "compressive learning," an approach to large-scale machine learning where datasets are massively compressed before learning (e.g., clustering, classification, or regression) is performed. In particular, a "sketch" is…

Encoding a sequence of observations is an essential task with many applications. The encoding can become highly efficient when the observations are generated by a dynamical system. A dynamical system imposes regularities on the observations…

Machine Learning · Statistics 2018-05-29 Arash Mehrjou , Friedrich Solowjow , Sebastian Trimpe , Bernhard Schölkopf