Noise-induced degeneration in online learning

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Sato, Yuzuru, Tsutsui, Daiji, Fujiwara, Akio
Format: Preprint
Veröffentlicht: 2020
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913349230395392
author Sato, Yuzuru
Tsutsui, Daiji
Fujiwara, Akio
author_facet Sato, Yuzuru
Tsutsui, Daiji
Fujiwara, Akio
contents In order to elucidate the plateau phenomena caused by vanishing gradient, we herein analyse stability of stochastic gradient descent near degenerated subspaces in a multi-layer perceptron. In stochastic gradient descent for Fukumizu-Amari model, which is the minimal multi-layer perceptron showing non-trivial plateau phenomena, we show that (1) attracting regions exist in multiply degenerated subspaces, (2) a strong plateau phenomenon emerges as a noise-induced synchronisation, which is not observed in deterministic gradient descent, (3) an optimal fluctuation exists to minimise the escape time from the degenerated subspace. The noise-induced degeneration observed herein is expected to be found in a broad class of machine learning via neural networks.
format Preprint
id arxiv_https___arxiv_org_abs_2008_10498
institution arXiv
publishDate 2020
record_format arxiv
spellingShingle Noise-induced degeneration in online learning
Sato, Yuzuru
Tsutsui, Daiji
Fujiwara, Akio
Adaptation and Self-Organizing Systems
Machine Learning
Dynamical Systems
Chaotic Dynamics
In order to elucidate the plateau phenomena caused by vanishing gradient, we herein analyse stability of stochastic gradient descent near degenerated subspaces in a multi-layer perceptron. In stochastic gradient descent for Fukumizu-Amari model, which is the minimal multi-layer perceptron showing non-trivial plateau phenomena, we show that (1) attracting regions exist in multiply degenerated subspaces, (2) a strong plateau phenomenon emerges as a noise-induced synchronisation, which is not observed in deterministic gradient descent, (3) an optimal fluctuation exists to minimise the escape time from the degenerated subspace. The noise-induced degeneration observed herein is expected to be found in a broad class of machine learning via neural networks.
title Noise-induced degeneration in online learning
topic Adaptation and Self-Organizing Systems
Machine Learning
Dynamical Systems
Chaotic Dynamics
url https://arxiv.org/abs/2008.10498