Reconstructing shared dynamics with a deep neural network

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Benkő, Zsigmond, Somogyvári, Zoltán
Format: Preprint
Published: 2021
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916503286185984
author Benkő, Zsigmond
Somogyvári, Zoltán
author_facet Benkő, Zsigmond
Somogyvári, Zoltán
contents Determining hidden shared patterns behind dynamic phenomena can be a game-changer in multiple areas of research. Here we present the principles and show a method to identify hidden shared dynamics from time series by a two-module, feedforward neural network architecture: the Mapper-Coach network. We reconstruct unobserved, continuous latent variable input, the time series generated by a chaotic logistic map, from the observed values of two simultaneously forced chaotic logistic maps. The network has been trained to predict one of the observed time series based on its own past and conditioned on the other observed time series by error-back propagation. It was shown, that after this prediction have been learned successfully, the activity of the bottleneck neuron, connecting the mapper and the coach module, correlated strongly with the latent shared input variable. The method has the potential to reveal hidden components of dynamical systems, where experimental intervention is not possible.
format Preprint
id arxiv_https___arxiv_org_abs_2105_02322
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle Reconstructing shared dynamics with a deep neural network
Benkő, Zsigmond
Somogyvári, Zoltán
Neural and Evolutionary Computing
68T07
Determining hidden shared patterns behind dynamic phenomena can be a game-changer in multiple areas of research. Here we present the principles and show a method to identify hidden shared dynamics from time series by a two-module, feedforward neural network architecture: the Mapper-Coach network. We reconstruct unobserved, continuous latent variable input, the time series generated by a chaotic logistic map, from the observed values of two simultaneously forced chaotic logistic maps. The network has been trained to predict one of the observed time series based on its own past and conditioned on the other observed time series by error-back propagation. It was shown, that after this prediction have been learned successfully, the activity of the bottleneck neuron, connecting the mapper and the coach module, correlated strongly with the latent shared input variable. The method has the potential to reveal hidden components of dynamical systems, where experimental intervention is not possible.
title Reconstructing shared dynamics with a deep neural network
topic Neural and Evolutionary Computing
68T07
url https://arxiv.org/abs/2105.02322