Recent Advances in Non-convex Smoothness Conditions and Applicability to Deep Linear Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Patel, Vivak, Varner, Christian
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909321274589184
author Patel, Vivak
Varner, Christian
author_facet Patel, Vivak
Varner, Christian
contents The presence of non-convexity in smooth optimization problems arising from deep learning have sparked new smoothness conditions in the literature and corresponding convergence analyses. We discuss these smoothness conditions, order them, provide conditions for determining whether they hold, and evaluate their applicability to training a deep linear neural network for binary classification.
format Preprint
id arxiv_https___arxiv_org_abs_2409_13672
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Recent Advances in Non-convex Smoothness Conditions and Applicability to Deep Linear Neural Networks
Patel, Vivak
Varner, Christian
Machine Learning
Optimization and Control
65K10, 68T07
The presence of non-convexity in smooth optimization problems arising from deep learning have sparked new smoothness conditions in the literature and corresponding convergence analyses. We discuss these smoothness conditions, order them, provide conditions for determining whether they hold, and evaluate their applicability to training a deep linear neural network for binary classification.
title Recent Advances in Non-convex Smoothness Conditions and Applicability to Deep Linear Neural Networks
topic Machine Learning
Optimization and Control
65K10, 68T07
url https://arxiv.org/abs/2409.13672