中文
相关论文

相关论文: The Effects of Mild Over-parameterization on the O…

200 篇论文

While over-parameterization is widely believed to be crucial for the success of optimization for the neural networks, most existing theories on over-parameterization do not fully explain the reason -- they either work in the Neural Tangent…

机器学习 · 计算机科学 2021-07-06 Mo Zhou , Rong Ge , Chi Jin

We theoretically study the landscape of the training error for neural networks in overparameterized cases. We consider three basic methods for embedding a network into a wider one with more hidden units, and discuss whether a minimum point…

机器学习 · 计算机科学 2019-06-17 Kenji Fukumizu , Shoichiro Yamaguchi , Yoh-ichi Mototake , Mirai Tanaka

Deep learning, in the form of artificial neural networks, has achieved remarkable practical success in recent years, for a variety of difficult machine learning applications. However, a theoretical explanation for this remains a major open…

机器学习 · 计算机科学 2016-06-15 Itay Safran , Ohad Shamir

We revisit the problem of learning a single neuron with ReLU activation under Gaussian input with square loss. We particularly focus on the over-parameterization setting where the student network has $n\ge 2$ neurons. We prove the global…

机器学习 · 计算机科学 2023-10-11 Weihang Xu , Simon S. Du

Recent theoretical works on over-parameterized neural nets have focused on two aspects: optimization and generalization. Many existing works that study optimization and generalization together are based on neural tangent kernel and require…

机器学习 · 计算机科学 2021-04-27 Shiyu Liang , Ruoyu Sun , R. Srikant

Overparameterization refers to the important phenomenon where the width of a neural network is chosen such that learning algorithms can provably attain zero loss in nonconvex training. The existing theory establishes such global convergence…

机器学习 · 计算机科学 2021-11-04 Chaehwan Song , Ali Ramezani-Kebrya , Thomas Pethick , Armin Eftekhari , Volkan Cevher

Many modern neural network architectures are trained in an overparameterized regime where the parameters of the model exceed the size of the training dataset. Sufficiently overparameterized neural network architectures in principle have the…

机器学习 · 计算机科学 2019-02-14 Samet Oymak , Mahdi Soltanolkotabi

We study the loss landscape of both shallow and deep, mildly overparameterized ReLU neural networks on a generic finite input dataset for the squared error loss. We show both by count and volume that most activation patterns correspond to…

机器学习 · 计算机科学 2024-02-09 Kedar Karhadkar , Michael Murray , Hanna Tseran , Guido Montúfar

We develop a convex analytic approach to analyze finite width two-layer ReLU networks. We first prove that an optimal solution to the regularized training problem can be characterized as extreme points of a convex set, where simple…

机器学习 · 计算机科学 2021-09-01 Tolga Ergen , Mert Pilanci

A recent line of work has shown that an overparametrized neural network can perfectly fit the training data, an otherwise often intractable nonconvex optimization problem. For (fully-connected) shallow networks, in the best case scenario,…

机器学习 · 计算机科学 2019-10-30 Armin Eftekhari , ChaeHwan Song , Volkan Cevher

One of the mysteries in the success of neural networks is randomly initialized first order methods like gradient descent can achieve zero training loss even though the objective function is non-convex and non-smooth. This paper demystifies…

机器学习 · 计算机科学 2019-02-06 Simon S. Du , Xiyu Zhai , Barnabas Poczos , Aarti Singh

The success of deep learning is due, to a large extent, to the remarkable effectiveness of gradient-based optimization methods applied to large neural networks. The purpose of this work is to propose a modern view and a general mathematical…

机器学习 · 计算机科学 2021-05-28 Chaoyue Liu , Libin Zhu , Mikhail Belkin

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…

机器学习 · 计算机科学 2018-06-12 Sanjeev Arora , Nadav Cohen , Elad Hazan

The optimization of neural networks under weight decay remains poorly understood from a theoretical standpoint. While weight decay is standard practice in modern training procedures, most theoretical analyses focus on unregularized…

机器学习 · 计算机科学 2025-05-29 Etienne Boursier , Matthew Bowditch , Matthias Englert , Ranko Lazic

Despite extensive studies, the underlying reason as to why overparameterized neural networks can generalize remains elusive. Existing theory shows that common stochastic optimizers prefer flatter minimizers of the training loss, and thus a…

机器学习 · 计算机科学 2023-07-25 Kaiyue Wen , Zhiyuan Li , Tengyu Ma

Understanding generalization of overparametrized neural networks remains a fundamental challenge in machine learning. Most of the literature mostly studies generalization from an interpolation point of view, taking convergence of parameters…

机器学习 · 计算机科学 2025-06-13 Etienne Boursier , Nicolas Flammarion

We develop an analytical framework to characterize the set of optimal ReLU neural networks by reformulating the non-convex training problem as a convex program. We show that the global optima of the convex parameterization are given by a…

机器学习 · 计算机科学 2024-01-22 Aaron Mishkin , Mert Pilanci

The analysis of neural network training beyond their linearization regime remains an outstanding open question, even in the simplest setup of a single hidden-layer. The limit of infinitely wide networks provides an appealing route forward…

机器学习 · 计算机科学 2020-06-19 Jaume de Dios , Joan Bruna

We study the population loss landscape of two-layer ReLU networks of the form $\sum_{k=1}^K \mathrm{ReLU}(w_k^\top x)$ in a realisable teacher-student setting with Gaussian covariates. We show that local minima admit an exact…

机器学习 · 统计学 2026-04-13 Jie Huang , Bruno Loureiro , Stefano Sarao Mannelli

We consider the optimization problem associated with training simple ReLU neural networks of the form $\mathbf{x}\mapsto \sum_{i=1}^{k}\max\{0,\mathbf{w}_i^\top \mathbf{x}\}$ with respect to the squared loss. We provide a computer-assisted…

机器学习 · 计算机科学 2018-08-10 Itay Safran , Ohad Shamir
‹ 上一页 1 2 3 10 下一页 ›