English
Related papers

Related papers: The Implicit Bias of Depth: How Incremental Learni…

200 papers

We perform an average case analysis of the generalization dynamics of large neural networks trained using gradient descent. We study the practically-relevant "high-dimensional" regime where the number of free parameters in the network is on…

Machine Learning · Statistics 2017-10-11 Madhu S. Advani , Andrew M. Saxe

Conventional wisdom in deep learning states that increasing depth improves expressiveness but complicates optimization. This paper suggests that, sometimes, increasing depth can speed up optimization. The effect of depth on optimization is…

Machine Learning · Computer Science 2018-06-12 Sanjeev Arora , Nadav Cohen , Elad Hazan

Understanding the implicit bias of training algorithms is of crucial importance in order to explain the success of overparametrised neural networks. In this paper, we study the dynamics of stochastic gradient descent over diagonal linear…

Machine Learning · Computer Science 2021-12-08 Scott Pesme , Loucas Pillaud-Vivien , Nicolas Flammarion

Deep Neural Networks are well known for efficiently fitting training data, yet experiencing poor generalization capabilities whenever some kind of bias dominates over the actual task labels, resulting in models learning "shortcuts". In…

Machine Learning · Computer Science 2024-08-12 Pietro Morerio , Ruggero Ragonesi , Vittorio Murino

Deep learning achieves remarkable generalization capability with overwhelming number of model parameters. Theoretical understanding of deep learning generalization receives recent attention yet remains not fully explored. This paper…

Machine Learning · Computer Science 2017-11-22 Guanhua Zheng , Jitao Sang , Changsheng Xu

Regardless of the marked differences between biological and artificial neural systems, one fundamental similarity is that they are essentially dynamical systems that can learn to imitate other dynamical systems, without knowing their…

Neurons and Cognition · Quantitative Biology 2019-11-06 Zhixin Lu , Danielle S. Bassett

Purpose: We propose a novel method for continual learning based on the increasing depth of neural networks. This work explores whether extending neural network depth may be beneficial in a life-long learning setting. Methods: We propose a…

Machine Learning · Computer Science 2023-05-09 Jędrzej Kozal , Michał Woźniak

When training deep neural networks with gradient descent, sharpness often increases -- a phenomenon known as progressive sharpening -- before saturating at the edge of stability. Although commonly observed in practice, the underlying…

Machine Learning · Computer Science 2025-06-10 Geonhui Yoo , Minhak Song , Chulhee Yun

We present experiments demonstrating that some other form of capacity control, different from network size, plays a central role in learning multilayer feed-forward networks. We argue, partially through analogy to matrix factorization, that…

Machine Learning · Computer Science 2015-04-17 Behnam Neyshabur , Ryota Tomioka , Nathan Srebro

A main puzzle of deep networks revolves around the absence of overfitting despite large overparametrization and despite the large capacity demonstrated by zero training error on randomly labeled data. In this note, we show that the dynamics…

A deep equilibrium model uses implicit layers, which are implicitly defined through an equilibrium point of an infinite sequence of computation. It avoids any explicit computation of the infinite sequence by finding an equilibrium point…

Machine Learning · Computer Science 2021-02-19 Kenji Kawaguchi

We identify and formalize a fundamental gradient descent phenomenon resulting in a learning proclivity in over-parameterized neural networks. Gradient Starvation arises when cross-entropy loss is minimized by capturing only a subset of…

Machine Learning · Computer Science 2021-11-25 Mohammad Pezeshki , Sékou-Oumar Kaba , Yoshua Bengio , Aaron Courville , Doina Precup , Guillaume Lajoie

We theoretically characterize gradient descent dynamics in deep linear networks trained at large width from random initialization and on large quantities of random data. Our theory captures the ``wider is better" effect of…

Machine Learning · Computer Science 2025-06-17 Blake Bordelon , Cengiz Pehlevan

Deep learning has arguably achieved tremendous success in recent years. In simple words, deep learning uses the composition of many nonlinear functions to model the complex dependency between input features and labels. While neural networks…

Machine Learning · Statistics 2019-04-16 Jianqing Fan , Cong Ma , Yiqiao Zhong

We conjecture that the inherent difference in generalisation between adaptive and non-adaptive gradient methods in deep learning stems from the increased estimation noise in the flattest directions of the true loss surface. We demonstrate…

Machine Learning · Statistics 2022-03-17 Diego Granziol , Nicholas Baskerville

Deep neural networks achieve stellar generalisation on a variety of problems, despite often being large enough to easily fit all their training data. Here we study the generalisation dynamics of two-layer neural networks in a…

Machine Learning · Statistics 2019-06-21 Sebastian Goldt , Madhu S. Advani , Andrew M. Saxe , Florent Krzakala , Lenka Zdeborová

Conventional wisdom attributes the mysterious generalization abilities of overparameterized neural networks to gradient descent (and its variants). The recent volume hypothesis challenges this view: it posits that these generalization…

Machine Learning · Computer Science 2025-12-19 Yotam Alexander , Yonatan Slutzky , Yuval Ran-Milo , Nadav Cohen

Neural networks trained via gradient descent with random initialization and without any regularization enjoy good generalization performance in practice despite being highly overparametrized. A promising direction to explain this phenomenon…

Machine Learning · Computer Science 2022-05-17 Hancheng Min , Salma Tarmoun , Rene Vidal , Enrique Mallada

Implicit deep learning prediction rules generalize the recursive rules of feedforward neural networks. Such rules are based on the solution of a fixed-point equation involving a single vector of hidden features, which is thus only…

Machine Learning · Computer Science 2020-08-10 Laurent El Ghaoui , Fangda Gu , Bertrand Travacca , Armin Askari , Alicia Y. Tsai

Continual learning, the ability of a model to adapt to an ongoing sequence of tasks without forgetting earlier ones, is a central goal of artificial intelligence. To better understand its underlying mechanisms, we study the limitations of…

Machine Learning · Statistics 2026-04-21 Hossein Taheri , Avishek Ghosh , Arya Mazumdar