中文
相关论文

相关论文: The Dynamics of Gradient Descent for Overparametri…

200 篇论文

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…

机器学习 · 计算机科学 2022-05-17 Hancheng Min , Salma Tarmoun , Rene Vidal , Enrique Mallada

The main aim of this paper is to provide an analysis of gradient descent (GD) algorithms with gradient errors that do not necessarily vanish, asymptotically. In particular, sufficient conditions are presented for both stability (almost sure…

系统与控制 · 计算机科学 2017-09-19 Arunselvan Ramaswamy , Shalabh Bhatnagar

The paper contains approximation guarantees for neural networks that are trained with gradient flow, with error measured in the continuous $L_2(\mathbb{S}^{d-1})$-norm on the $d$-dimensional unit sphere and targets that are Sobolev smooth.…

机器学习 · 计算机科学 2023-09-12 G. Welper

Neural networks typically generalize well when fitting the data perfectly, even though they are heavily overparameterized. Many factors have been pointed out as the reason for this phenomenon, including an implicit bias of stochastic…

机器学习 · 计算机科学 2025-02-04 Amit Peleg , Matthias Hein

Implicit deep learning has received increasing attention recently due to the fact that it generalizes the recursive prediction rules of many commonly used neural network architectures. Its prediction rule is provided implicitly based on the…

机器学习 · 计算机科学 2022-02-21 Tianxiang Gao , Hailiang Liu , Jia Liu , Hridesh Rajan , Hongyang Gao

Stochastic gradient descent (SGD) has been found to be surprisingly effective in training a variety of deep neural networks. However, there is still a lack of understanding on how and why SGD can train these complex networks towards a…

机器学习 · 计算机科学 2019-01-03 Yi Zhou , Junjie Yang , Huishuai Zhang , Yingbin Liang , Vahid Tarokh

Theoretically understanding stochastic gradient descent (SGD) in overparameterized models has led to the development of several optimization algorithms that are widely used in practice today. Recent work by~\citet{zou2021benign} provides…

机器学习 · 计算机科学 2025-06-19 Alexandru Meterez , Depen Morwani , Costin-Andrei Oncescu , Jingfeng Wu , Cengiz Pehlevan , Sham Kakade

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…

机器学习 · 计算机科学 2023-05-30 Mingze Wang , Chao Ma

We study learning to learn for regression problems through the lens of hyperparameter tuning. We propose the Langevin Gradient Descent Algorithm (LGD), which approximates the mean of the posterior distribution defined by the loss function…

机器学习 · 计算机科学 2026-04-16 Saumya Goyal , Rohith Rongali , Ritabrata Ray , Barnabás Póczos

A main puzzle of deep neural networks (DNNs) revolves around the apparent absence of "overfitting", defined in this paper as follows: the expected error does not get worse when increasing the number of neurons or of iterations of gradient…

机器学习 · 计算机科学 2018-07-02 Tomaso Poggio , Qianli Liao , Brando Miranda , Andrzej Banburski , Xavier Boix , Jack Hidary

Overparametrized neural networks trained by gradient descent (GD) can provably overfit any training data. However, the generalization guarantee may not hold for noisy data. From a nonparametric perspective, this paper studies how well…

机器学习 · 统计学 2021-09-28 Tianyang Hu , Wenjia Wang , Cong Lin , Guang Cheng

Overparameterized ML models, including neural networks, typically induce underdetermined training objectives with multiple global minima. The implicit bias refers to the limiting global minimum that is attained by a common optimization…

机器学习 · 统计学 2026-03-06 Kuo-Wei Lai , Guanghui Wang , Molei Tao , Vidya Muthukumar

Machine learning methods are commonly used to solve inverse problems, wherein an unknown signal must be estimated from few indirect measurements generated via a known acquisition procedure. In particular, neural networks perform well…

机器学习 · 计算机科学 2025-12-05 Hannah Laus , Suzanna Parkinson , Vasileios Charisopoulos , Felix Krahmer , Rebecca Willett

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…

机器学习 · 统计学 2017-10-11 Madhu S. Advani , Andrew M. Saxe

Even for the gradient descent (GD) method applied to neural network training, understanding its optimization dynamics, including convergence rate, iterate trajectories, function value oscillations, and especially its implicit acceleration,…

机器学习 · 计算机科学 2026-05-22 Alexander Tyurin

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…

机器学习 · 计算机科学 2019-12-06 Gauthier Gidel , Francis Bach , Simon Lacoste-Julien

We study the convergence dynamics of Gradient Descent (GD) in a minimal binary classification setting, consisting of a two-neuron ReLU network and two training instances. We prove that even under these strong simplifying assumptions, while…

机器学习 · 计算机科学 2026-03-03 Guy Smorodinsky , Sveta Gimpleson , Itay Safran

This paper presents a comprehensive study on the convergence rates of the stochastic gradient descent (SGD) algorithm when applied to overparameterized two-layer neural networks. Our approach combines the Neural Tangent Kernel (NTK)…

机器学习 · 统计学 2024-07-11 Dinghao Cao , Zheng-Chu Guo , Lei Shi

Normalized gradient descent has shown substantial success in speeding up the convergence of exponentially-tailed loss functions (which includes exponential and logistic losses) on linear classifiers with separable data. In this paper, we go…

机器学习 · 计算机科学 2023-06-28 Hossein Taheri , Christos Thrampoulidis

We demonstrate that applying an eventual decay to the learning rate (LR) in empirical risk minimization (ERM), where the mean-squared-error loss is minimized using standard gradient descent (GD) for training a two-layer neural network with…

机器学习 · 统计学 2026-02-10 Kyle Sung , Kholood Khalil , Noah Forman , Steven Samu , Anastasis Kratsios