English
Related papers

Related papers: Convergence of gradient descent for learning linea…

200 papers

We show that gradient descent converges to a local minimizer, almost surely with random initialization. This is proved by applying the Stable Manifold Theorem from dynamical systems theory.

Machine Learning · Statistics 2016-03-07 Jason D. Lee , Max Simchowitz , Michael I. Jordan , Benjamin Recht

We consider the minimization of non-convex quadratic forms regularized by a cubic term, which exhibit multiple saddle points and poor local minima. Nonetheless, we prove that, under mild assumptions, gradient descent approximates the…

Optimization and Control · Mathematics 2022-08-31 Yair Carmon , John C. Duchi

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.…

Machine Learning · Computer Science 2023-09-12 G. Welper

Gradient descent, or negative gradient flow, is a standard technique in optimization to find minima of functions. Many implementations of gradient descent rely on discretized versions, i.e., moving in the gradient direction for a set step…

Differential Geometry · Mathematics 2024-07-01 Dara Gold , Steven Rosenberg

We study the training dynamics of gradient descent in a softmax self-attention layer trained to perform linear regression and show that a simple first-order optimization algorithm can converge to the globally optimal self-attention…

Machine Learning · Computer Science 2026-03-03 Gautam Goel , Mahdi Soltanolkotabi , Peter Bartlett

We propose a tensor-based criterion for benign landscape in phase retrieval and establish boundedness of gradient trajectories. This implies that gradient descent will converge to a global minimum for almost every initial point.

Optimization and Control · Mathematics 2024-10-18 Théodore Fougereux , Cédric Josz , Xiaopeng Li

Empirically it has been observed that the performance of deep neural networks steadily improves as we increase model size, contradicting the classical view on overfitting and generalization. Recently, the double descent phenomena has been…

Machine Learning · Computer Science 2021-07-28 Ilja Kuzborskij , Csaba Szepesvári , Omar Rivasplata , Amal Rannen-Triki , Razvan Pascanu

The simplicity of gradient descent (GD) made it the default method for training ever-deeper and complex neural networks. Both loss functions and architectures are often explicitly tuned to be amenable to this basic local optimization. In…

Machine Learning · Computer Science 2019-04-30 Dmitrii Marin , Meng Tang , Ismail Ben Ayed , Yuri Boykov

We examine gradient descent on unregularized logistic regression problems, with homogeneous linear predictors on linearly separable datasets. We show the predictor converges to the direction of the max-margin (hard margin SVM) solution. The…

Machine Learning · Statistics 2024-10-29 Daniel Soudry , Elad Hoffer , Mor Shpigel Nacson , Suriya Gunasekar , Nathan Srebro

In spite of the accomplishments of deep learning based algorithms in numerous applications and very broad corresponding research interest, at the moment there is still no rigorous understanding of the reasons why such algorithms produce…

Statistics Theory · Mathematics 2020-03-04 Arnulf Jentzen , Timo Welti

Traditional landscape analysis of deep neural networks aims to show that no sub-optimal local minima exist in some appropriate sense. From this, one may be tempted to conclude that descent algorithms which escape saddle points will reach a…

Machine Learning · Computer Science 2020-01-01 Shiyu Liang , Ruoyu Sun , R. Srikant

Gradient descent is the primary workhorse for optimizing large-scale problems in machine learning. However, its performance is highly sensitive to the choice of the learning rate. A key limitation of gradient descent is its lack of natural…

Optimization and Control · Mathematics 2025-07-15 Oscar Smee , Fred Roosta , Stephen J. Wright

Stochastic gradient descent plays a fundamental role in nearly all applications of deep learning. However its ability to converge to a global minimum remains shrouded in mystery. In this paper we propose to study the behavior of the loss…

Machine Learning · Computer Science 2023-02-02 Mark Sandler , Andrey Zhmoginov , Max Vladymyrov , Nolan Miller

In this paper, we propose a simple, fast and easy to implement algorithm LOSSGRAD (locally optimal step-size in gradient descent), which automatically modifies the step-size in gradient descent during neural networks training. Given a…

Machine Learning · Computer Science 2019-11-26 Bartosz Wójcik , Łukasz Maziarka , Jacek Tabor

Deep neural networks (DNNs) have demonstrated dominating performance in many fields; since AlexNet, networks used in practice are going wider and deeper. On the theoretical side, a long line of works has been focusing on training neural…

Machine Learning · Computer Science 2019-06-18 Zeyuan Allen-Zhu , Yuanzhi Li , Zhao Song

We study the convergence of a variant of distributed gradient descent (DGD) on a distributed low-rank matrix approximation problem wherein some optimization variables are used for consensus (as in classical DGD) and some optimization…

Optimization and Control · Mathematics 2018-12-27 Zhihui Zhu , Qiuwei Li , Xinshuo Yang , Gongguo Tang , Michael B. Wakin

Due to the success of deep learning to solving a variety of challenging machine learning tasks, there is a rising interest in understanding loss functions for training neural networks from a theoretical aspect. Particularly, the properties…

Machine Learning · Statistics 2017-11-01 Yi Zhou , Yingbin Liang

Optimization problems with rank constraints arise in many applications, including matrix regression, structured PCA, matrix completion and matrix decomposition problems. An attractive heuristic for solving such problems is to factorize the…

Statistics Theory · Mathematics 2015-09-11 Yudong Chen , Martin J. Wainwright

Recent success in training deep neural networks have prompted active investigation into the features learned on their intermediate layers. Such research is difficult because it requires making sense of non-linear computations performed by…

Machine Learning · Computer Science 2016-03-01 Yixuan Li , Jason Yosinski , Jeff Clune , Hod Lipson , John Hopcroft

Herein the topics of (natural) gradient descent, data decorrelation, and approximate methods for backpropagation are brought into a common discussion. Natural gradient descent illuminates how gradient vectors, pointing at directions of…

Machine Learning · Computer Science 2025-08-26 Nasir Ahmad