☆ 4.6 Article

Training neural networks using Metropolis Monte Carlo and an adaptive variant

MACHINE LEARNING-SCIENCE AND TECHNOLOGY (2022)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Multidisciplinary Sciences

Correspondence between neuroevolution and gradient descent

Stephen Whitelam et al.

Summary: Authors derive an analytic equivalence between neural network training under conditioned stochastic mutations and under gradient descent, showing that in the presence of small mutations, training a neural network by conditioned stochastic mutation or neuroevolution of its weights is equivalent to gradient descent on the loss function with Gaussian white noise. Neuroevolution is found to be equivalent to gradient descent on the loss function when averaged over independent realizations of the learning process, which is demonstrated through numerical simulations across finite mutations and various neural network architectures. This provides a connection between two families of neural-network training methods that are usually considered to be fundamentally different.

NATURE COMMUNICATIONS (2021)

添加到收藏夹

Article Physics, Condensed Matter

Statistical Mechanics of Deep Learning

Yasaman Bahri et al.

Annual Review of Condensed Matter Physics (2019)

添加到收藏夹

Review Multidisciplinary Sciences

Deep learning

Yann LeCun et al.

NATURE (2015)

添加到收藏夹

Review Computer Science, Artificial Intelligence

Deep learning in neural networks: An overview

Juergen Schmidhuber

NEURAL NETWORKS (2015)

添加到收藏夹

Proceedings Paper Computer Science, Information Systems

Simulated Annealing Algorithm for Deep Learning

L. M. Rasdi Rere et al.

THIRD INFORMATION SYSTEMS INTERNATIONAL CONFERENCE 2015 (2015)

添加到收藏夹

Article Mathematics, Applied

Recurrent policy gradients

Daan Wierstra et al.

LOGIC JOURNAL OF THE IGPL (2010)

添加到收藏夹

Article Chemistry, Physical

Avoiding unphysical kinetic traps in Monte Carlo simulations of strongly attractive particles

Stephen Whitelam et al.

JOURNAL OF CHEMICAL PHYSICS (2007)

添加到收藏夹

Article Physics, Fluids & Plasmas

Marshall rosenbluth and the metropolis algorithm

JE Gubernatis

PHYSICS OF PLASMAS (2005)

添加到收藏夹

Article Physics, Multidisciplinary

Rejection-free geometric cluster algorithm for complex fluids

JW Liu et al.

PHYSICAL REVIEW LETTERS (2004)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Reducing the time complexity of the derandomized evolution strategy with covariance matrix adaptation (CMA-ES)

N Hansen et al.

EVOLUTIONARY COMPUTATION (2003)

添加到收藏夹

Article Chemistry, Physical

Improving the efficiency of the aggregation-volume-bias Monte Carlo algorithm

B Chen et al.

JOURNAL OF PHYSICAL CHEMISTRY B (2001)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Completely derandomized self-adaptation in evolution strategies

N Hansen et al.

EVOLUTIONARY COMPUTATION (2001)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.