Global Minima of Overparameterized Neural Networks

DOI10.1137/19M1308943zbMath1483.68288arXiv1804.10200OpenAlexW3160443277MaRDI QIDQ4999400

Yaim Cooper

Publication date: 6 July 2021

Published in: SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1804.10200

zbMATH Keywords

manifoldoptimizationeigenvalues of the Hessianoverparameterizationlandscape

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Related Items (4)

Loss landscapes and optimization in over-parameterized non-linear systems and neural networks ⋮ On the Benefit of Width for Neural Networks: Disappearance of Basins ⋮ Central limit theorems for stochastic gradient descent with averaging for stable manifolds ⋮ Landscape and training regimes in deep learning

Uses Software

• Entropy-SGD

Cites Work

• Introduction to Smooth Manifolds
• Entropy-SGD: biasing gradient descent into wide valleys
• Unnamed Item
• Unnamed Item

This page was built for publication: Global Minima of Overparameterized Neural Networks