paper-with-me

Papers

Gradient Descent with Random Initialization: Fast Global Convergence for Nonconvex Phase Retrieval

2018-03-21 · Yuxin Chen, Yuejie Chi, Jianqing Fan, Cong Ma

This paper considers the problem of solving systems of quadratic equations, namely, recovering an object of interest $\mathbf{x}^{\natural}\in\mathbb{R}^{n}$ from $m$ quadratic equations/samples $y_{i}=(\mathbf{a}_{i}^{\top}\mathbf{x}^{\natural})^{2}$, $1\leq i\leq m$. This problem, also dubbed as phase retrieval, spans multiple domains including physical sciences and machine learning. We investigate the efficiency of gradient descent (or Wirtinger flow) designed for the nonconvex least squares problem. We prove that under Gaussian designs, gradient descent --- when randomly initialized --- yields an $\epsilon$-accurate solution in $O\big(\log n+\log(1/\epsilon)\big)$ iterations given nearly minimal samples, thus achieving near-optimal computational and sample complexities at once. This provides the first global convergence guarantee concerning vanilla gradient descent for phase retrieval, without the need of (i) carefully-designed initialization, (ii) sample splitting, or (iii) sophisticated saddle-point escaping schemes. All of these are achieved by exploiting the statistical models in analyzing optimization algorithms, via a leave-one-out approach that enables the decoupling of certain statistical dependency between the gradient descent iterates and the data.

📄 PDF Abstract BibTeX arXiv:1803.07726

Code (0)

등록된 구현이 없습니다.

Tasks

Retrieval

Similar Papers 제목 키워드 기반

Preconditioning Matters: Fast Global Convergence of Non-convex Matrix Factorization via Scaled Gradient Descent

2023-09-21 · NeurIPS 2023 11

Low-rank matrix factorization (LRMF) is a canonical problem in non-convex optimization, the objective function to be minimized is non-convex and even non-smooth, which makes the global convergence guarantee of gradient-b…

Fast Global Convergence for Low-rank Matrix Recovery via Riemannian Gradient Descent with Random Initialization

2020-12-31 · Thomas Y. Hou, Zhenzhen Li, Ziyun Zhang

In this paper, we propose a new global analysis framework for a class of low-rank matrix recovery problems on the Riemannian manifold. We analyze the global behavior for the Riemannian optimization with random initializa…

RetrievalRiemannian optimization

Stochastic Gradient Descent Optimizes Over-parameterized Deep ReLU Networks

2018-11-21 · Difan Zou, Yuan Cao, Dongruo Zhou, Quanquan Gu

We study the problem of training deep neural networks with Rectified Linear Unit (ReLU) activation function using gradient descent and stochastic gradient descent. In particular, we study the binary classification proble…

Binary Classification

Gradient Descent Provably Optimizes Over-parameterized Neural Networks

2018-10-04 · ICLR 2019 5 · Simon S. Du, Xiyu Zhai, Barnabas Poczos, Aarti Singh

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. Thi…

Global Convergence of Four-Layer Matrix Factorization under Random Initialization

2025-11-13 · Minrui Luo, Weihang Xu, Xiang Gao, Maryam Fazel 외 arxiv

Gradient descent dynamics on the deep matrix factorization problem is extensively studied as a simplified theoretical model for deep neural networks. Although the convergence theory for two-layer matrix factorization is …