Work

Toward Moderate Overparameterization: Global Convergence Guarantees for Training Shallow Neural Networks

Year: 2020

Type: article

Source: IEEE Journal on Selected Areas in Information Theory

DOI: 10.1109/jsait.2020.2991332

Authors: Samet Oymak, Mahdi Soltanolkotabi

Institutions: University of California, Riverside, University of Southern California

Language: English

Topic: Stochastic Gradient Optimization Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: closed

Funders: National Science Foundation, David and Lucile Packard Foundation, Google, Defense Advanced Research Projects Agency, National Science Foundation of Sri Lanka, Air Force Office of Scientific Research

Awards: G2520208575, CAREER: Guaranteed Nonconvex Optimization for High-Dimensional Learning, G433588577, CIF: Small: Precise Computational and Statistical Tradeoffs for Iterative Signal Estimation and Supervised Learning

OpenAlex ID: W3021189130

FWCI: 9.38

Cites: 57

Cited by: 147

Related to: 10