Multiplex visibility graphs to investigate recurrent neural network dynamics.
Sci Rep
; 7: 44037, 2017 03 10.
Article
em En
| MEDLINE
| ID: mdl-28281563
ABSTRACT
A recurrent neural network (RNN) is a universal approximator of dynamical systems, whose performance often depends on sensitive hyperparameters. Tuning them properly may be difficult and, typically, based on a trial-and-error approach. In this work, we adopt a graph-based framework to interpret and characterize internal dynamics of a class of RNNs called echo state networks (ESNs). We design principled unsupervised methods to derive hyperparameters configurations yielding maximal ESN performance, expressed in terms of prediction error and memory capacity. In particular, we propose to model time series generated by each neuron activations with a horizontal visibility graph, whose topological properties have been shown to be related to the underlying system dynamics. Successively, horizontal visibility graphs associated with all neurons become layers of a larger structure called a multiplex. We show that topological properties of such a multiplex reflect important features of ESN dynamics that can be used to guide the tuning of its hyperparamers. Results obtained on several benchmarks and a real-world dataset of telephone call data records show the effectiveness of the proposed methods.
Texto completo:
1
Coleções:
01-internacional
Base de dados:
MEDLINE
Assunto principal:
Análise de Sistemas
/
Redes Neurais de Computação
/
Aprendizado de Máquina não Supervisionado
Tipo de estudo:
Prognostic_studies
Idioma:
En
Revista:
Sci Rep
Ano de publicação:
2017
Tipo de documento:
Article
País de afiliação:
Noruega