English
Related papers

Related papers: Phase Transitions between Accuracy Regimes in L2 r…

200 papers

We study the generalization behavior of transfer learning of deep neural networks (DNNs). We adopt the overparameterization perspective -- featuring interpolation of the training data (i.e., approximately zero train error) and the double…

Machine Learning · Computer Science 2023-06-13 Yehuda Dar , Lorenzo Luzi , Richard G. Baraniuk

When optimizing over-parameterized models, such as deep neural networks, a large set of parameters can achieve zero training error. In such cases, the choice of the optimization algorithm and its respective hyper-parameters introduces…

Machine Learning · Computer Science 2019-12-06 Gauthier Gidel , Francis Bach , Simon Lacoste-Julien

The convergence of GD and SGD when training mildly parameterized neural networks starting from random initialization is studied. For a broad range of models and loss functions, including the most commonly used square loss and cross entropy…

Machine Learning · Computer Science 2023-05-30 Mingze Wang , Chao Ma

Overparameterization in deep learning typically refers to settings where a trained neural network (NN) has representational capacity to fit the training data in many ways, some of which generalize well, while others do not. In the case of…

Machine Learning · Computer Science 2023-03-24 Edo Cohen-Karlik , Itamar Menuhin-Gruman , Raja Giryes , Nadav Cohen , Amir Globerson

Existing analyses of neural network training often operate under the unrealistic assumption of an extremely small learning rate. This lies in stark contrast to practical wisdom and empirical studies, such as the work of J. Cohen et al.…

Machine Learning · Computer Science 2023-10-20 Kwangjun Ahn , Sébastien Bubeck , Sinho Chewi , Yin Tat Lee , Felipe Suarez , Yi Zhang

We investigate the learning dynamics of fully-connected neural networks through the lens of gradient signal-to-noise ratio (SNR), examining the behavior of first-order optimizers like Adam in non-convex objectives. By interpreting the…

Machine Learning · Computer Science 2024-03-28 Sokratis J. Anagnostopoulos , Juan Diego Toscano , Nikolaos Stergiopulos , George Em Karniadakis

For high-dimensional Gaussian data, we investigate theoretically how the features of a two-layer neural network adapt to the structure of the target function through a few large batch gradient descent steps, leading to an improvement in the…

Machine Learning · Statistics 2025-06-05 Yatin Dandi , Florent Krzakala , Bruno Loureiro , Luca Pesce , Ludovic Stephan

The early phase of training of deep neural networks is critical for their final performance. In this work, we study how the hyperparameters of stochastic gradient descent (SGD) used in the early phase of training affect the rest of the…

Machine Learning · Computer Science 2020-02-25 Stanislaw Jastrzebski , Maciej Szymczak , Stanislav Fort , Devansh Arpit , Jacek Tabor , Kyunghyun Cho , Krzysztof Geras

Understanding the implicit regularization imposed by neural network architectures and gradient based optimization methods is a key challenge in deep learning and AI. In this work we provide sharp results for the implicit regularization…

Machine Learning · Computer Science 2023-07-17 Johan S. Wind , Vegard Antun , Anders C. Hansen

In this manuscript, we investigate the problem of how two-layer neural networks learn features from data, and improve over the kernel regime, after being trained with a single gradient descent step. Leveraging the insight from (Ba et al.,…

Machine Learning · Statistics 2024-09-06 Hugo Cui , Luca Pesce , Yatin Dandi , Florent Krzakala , Yue M. Lu , Lenka Zdeborová , Bruno Loureiro

Motivated by the gap between theoretical optimal approximation rates of deep neural networks (DNNs) and the accuracy realized in practice, we seek to improve the training of DNNs. The adoption of an adaptive basis viewpoint of DNNs leads to…

Machine Learning · Computer Science 2019-12-11 Eric C. Cyr , Mamikon A. Gulian , Ravi G. Patel , Mauro Perego , Nathaniel A. Trask

Modern neural networks are often operated in a strongly overparametrized regime: they comprise so many parameters that they can interpolate the training set, even if actual labels are replaced by purely random ones. Despite this, they…

Machine Learning · Statistics 2022-06-10 Andrea Montanari , Yiqiao Zhong

Supervised deep learning involves the training of neural networks with a large number $N$ of parameters. For large enough $N$, in the so-called over-parametrized regime, one can essentially fit the training data points. Sparsity-based…

Disordered Systems and Neural Networks · Physics 2020-04-22 Mario Geiger , Arthur Jacot , Stefano Spigler , Franck Gabriel , Levent Sagun , Stéphane d'Ascoli , Giulio Biroli , Clément Hongler , Matthieu Wyart

Background: It is still an open research area to theoretically understand why Deep Neural Networks (DNNs)---equipped with many more parameters than training data and trained by (stochastic) gradient-based methods---often achieve remarkably…

Machine Learning · Computer Science 2018-11-30 Zhiqin John Xu

The typical training of neural networks using large stepsize gradient descent (GD) under the logistic loss often involves two distinct phases, where the empirical risk oscillates in the first phase but decreases monotonically in the second…

Machine Learning · Statistics 2024-06-28 Yuhang Cai , Jingfeng Wu , Song Mei , Michael Lindsey , Peter L. Bartlett

This paper explores the connection between learning trajectories of Deep Neural Networks (DNNs) and their generalization capabilities when optimized using (stochastic) gradient descent algorithms. Instead of concentrating solely on the…

Machine Learning · Computer Science 2023-11-01 Jingwen Fu , Zhizheng Zhang , Dacheng Yin , Yan Lu , Nanning Zheng

A numerical and phenomenological study of the gradient descent (GD) algorithm for training two-layer neural network models is carried out for different parameter regimes when the target function can be accurately approximated by a…

Machine Learning · Computer Science 2020-06-26 Chao Ma , Lei Wu , Weinan E

Spectral bias, the tendency of neural networks to learn low frequencies first, can be both a blessing and a curse. While it enhances the generalization capabilities by suppressing high-frequency noise, it can be a limitation in scientific…

Machine Learning · Computer Science 2026-05-08 Shuai Jiang , Alexey Voronin , Eric Cyr , Ben Southworth

Over-parameterized models, such as large deep networks, often exhibit a double descent phenomenon, whereas a function of model size, error first decreases, increases, and decreases at last. This intriguing double descent behavior also…

Machine Learning · Computer Science 2020-09-22 Reinhard Heckel , Fatih Furkan Yilmaz

Deep learning experiments by Cohen et al. [2021] using deterministic Gradient Descent (GD) revealed an Edge of Stability (EoS) phase when learning rate (LR) and sharpness (i.e., the largest eigenvalue of Hessian) no longer behave as in…

Machine Learning · Computer Science 2022-10-31 Sanjeev Arora , Zhiyuan Li , Abhishek Panigrahi