Automatic, dynamic, and nearly optimal learning rate specification via local quadratic approximation

Authors:

Publisher website

1 September 2021

journal article
research article
Published by Elsevier in Neural Networks

Vol. 141, 11-29
DOI:10.1016/j.neunet.2021.03.025
PMID:33845311

Abstract

No abstract available

Keywords

All Related Versions

Version 1, 2020-04-07, ArXiv (Unconfirmed version)

Funding Information

National Natural Science Foundation of China (11701560, 12071477, 11525101, 11831008, 2020-XY-30, 71532001, 71873137)
Key Laboratory of Engineering Plastics, Chinese Academy of Sciences (11701560, 12071477, 11525101, 11831008, 2020-XY-30, 71532001, 71873137)

This publication has 15 references indexed in Scilit:

Human-level control through deep reinforcement learning
Nature, 2015
An optimal method for stochastic composite optimization
Mathematical Programming, 2011
An Incremental Gradient(-Projection) Method with Momentum Term and Adaptive Stepsize Rule
SIAM Journal on Optimization, 1998
Gradient-based learning applied to document recognition
Proceedings of the IEEE, 1998
Algorithm 778: L-BFGS-B
ACM Transactions on Mathematical Software, 1997
Historical Development of the Newton–Raphson Method
SIAM Review, 1995
Approximation capabilities of multilayer feedforward networks
Neural Networks, 1991
Backpropagation Applied to Handwritten Zip Code Recognition
Neural Computation, 1989
Learning representations by back-propagating errors
Nature, 1986
A Stochastic Approximation Method
The Annals of Mathematical Statistics, 1951