English
Related papers

Related papers: Egalitarian Gradient Descent: A Simple Approach to…

200 papers

We study grokking, the onset of generalization long after overfitting, in a classical ridge regression setting. We prove end-to-end grokking results for learning over-parameterized linear regression models using gradient descent with weight…

Machine Learning · Computer Science 2026-02-09 Mingyue Xu , Gal Vardi , Itay Safran

We study the dynamics of gradient flow with small weight decay on general training losses $F: \mathbb{R}^d \to \mathbb{R}$. Under mild regularity assumptions and assuming convergence of the unregularised gradient flow, we show that the…

Machine Learning · Computer Science 2025-11-06 Etienne Boursier , Scott Pesme , Radu-Alexandru Dragomir

We present a theoretical explanation of the ``grokking'' phenomenon, where a model generalizes long after overfitting,for the originally-studied problem of modular addition. First, we show that early in gradient descent, when the ``kernel…

Machine Learning · Computer Science 2024-07-18 Mohamad Amin Mohamadi , Zhiyuan Li , Lei Wu , Danica J. Sutherland

Neural network grokking -- the abrupt memorization-to-generalization transition -- challenges our understanding of learning dynamics. Through finite-size scaling of gradient avalanche dynamics across eight model scales, we find that…

Machine Learning · Computer Science 2026-04-07 Ping Wang

In continual learning problems, it is often necessary to overwrite components of a neural network's learned representation in response to changes in the data stream; however, neural networks often exhibit \primacy bias, whereby early…

Machine Learning · Computer Science 2025-07-29 Clare Lyle , Gharda Sokar , Razvan Pascanu , Andras Gyorgy

In some settings neural networks exhibit a phenomenon known as \textit{grokking}, where they achieve perfect or near-perfect accuracy on the validation set long after the same performance has been achieved on the training set. In this…

Machine Learning · Computer Science 2024-04-02 Jack Miller , Charles O'Neill , Thang Bui

We propose that the grokking phenomenon, where the train loss of a neural network decreases much earlier than its test loss, can arise due to a neural network transitioning from lazy training dynamics to a rich, feature learning regime. To…

Machine Learning · Statistics 2024-04-12 Tanishq Kumar , Blake Bordelon , Samuel J. Gershman , Cengiz Pehlevan

Training loss and accuracy are the standard signals used to monitor generalization during deep neural network training. Two well-documented phenomena complicate this picture: in grokking, train loss falls rapidly while test performance…

Machine Learning · Computer Science 2026-05-29 Chi-Ning Chou , Oscar Uzdelewicz , Neng-Chun Chiu , Yao-Yuan Yang , SueYeon Chung

Standard optimization theories struggle to explain grokking, where generalization occurs long after training convergence. While geometric studies attribute this to slow drift, they often overlook the interaction between the optimizer's…

Machine Learning · Computer Science 2026-03-17 Pratyush Acharya , Habish Dhakal

Adaptive Gradient Descent with Energy (AEGD) is a variant of gradient descent (GD) designed to mitigate step-size sensitivity through an energy-based formulation. AEGD is notable for its unconditional energy stability, which guarantees…

Optimization and Control · Mathematics 2025-12-16 Lin Feng , Hailiang Liu

Recent work by Power et al. (2022) highlighted a surprising "grokking" phenomenon in learning arithmetic tasks: a neural net first "memorizes" the training set, resulting in perfect training accuracy but near-random test accuracy, and after…

Machine Learning · Computer Science 2024-04-03 Kaifeng Lyu , Jikai Jin , Zhiyuan Li , Simon S. Du , Jason D. Lee , Wei Hu

Gradient regularization (GR) is a method that penalizes the gradient norm of the training loss during training. While some studies have reported that GR can improve generalization performance, little attention has been paid to it from the…

Machine Learning · Computer Science 2023-02-06 Ryo Karakida , Tomoumi Takase , Tomohiro Hayase , Kazuki Osawa

Optimization problem, which is aimed at finding the global minimal value of a given cost function, is one of the central problem in science and engineering. Various numerical methods have been proposed to solve this problem, among which the…

Optimization and Control · Mathematics 2022-10-07 Shaojun Dong , Fengyu Le , Meng Zhang , Si-Jing Tao , Chao Wang , Yong-Jian Han , Guo-Ping Guo

We present a simple neural network that can learn modular arithmetic tasks and exhibits a sudden jump in generalization known as ``grokking''. Concretely, we present (i) fully-connected two-layer networks that exhibit grokking on various…

Machine Learning · Computer Science 2023-01-10 Andrey Gromov

Gradient regularization (GR) has been shown to improve the generalizability of trained models. While Natural Gradient Descent has been shown to accelerate optimization in the initial phase of training, little attention has been paid to how…

Machine Learning · Computer Science 2026-03-27 Satya Prakash Dash , Hossein Abdi , Wei Pan , Samuel Kaski , Mingfei Sun

In this paper, we investigate the phenomenon of grokking, where models exhibit delayed generalization following overfitting on training data. We focus on data-scarce regimes where the number of training samples falls below the critical…

Machine Learning · Computer Science 2025-11-10 Vaibhav Singh , Eugene Belilovsky , Rahaf Aljundi

Grokking -- the delayed transition from memorization to generalization in small algorithmic tasks -- remains poorly understood. We present a geometric analysis of optimization dynamics in transformers trained on modular arithmetic. PCA of…

Machine Learning · Computer Science 2026-04-06 Yongzhong Xu

This paper focuses on predicting the occurrence of grokking in neural networks, a phenomenon in which perfect generalization emerges long after signs of overfitting or memorization are observed. It has been reported that grokking can only…

Machine Learning · Computer Science 2023-10-02 Pascal Jr. Tikeng Notsawo , Hattie Zhou , Mohammad Pezeshki , Irina Rish , Guillaume Dumas

Grokking is a phenomenon where a model trained on an algorithmic task first overfits but, then, after a large amount of additional training, undergoes a phase transition to generalize perfectly. We empirically study the internal structure…

Machine Learning · Computer Science 2023-03-22 William Merrill , Nikolaos Tsilivis , Aman Shukla

Grokking is the phenomenon where neural networks NNs initially fit the training data and later generalize to the test data during training. In this paper, we empirically provide a frequency perspective to explain the emergence of this…

Machine Learning · Computer Science 2024-05-29 Zhangchen Zhou , Yaoyu Zhang , Zhi-Qin John Xu