Analyzing controllability of neural networks

In recent years, due to the relation between cognitive control and mathematical concept of control dynamical systems, there has been growing interest in the descriptive analysis of complex networks with linear dynamics, permeating many aspects from everyday life, obtaining considerable advances in t...

Descripción completa

Detalles Bibliográficos
Autor: García Planas, María Isabel|||0000-0001-7418-7208
Tipo de recurso: artículo
Fecha de publicación:2019
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/129425
Acceso en línea:https://hdl.handle.net/2117/129425
Access Level:acceso abierto
Palabra clave:Neural networks (Computer science)
Neural network
controllability
exact controllability
eigenvalues
eigenvectors
linear systems
Xarxes neuronals (Informàtica)
Àrees temàtiques de la UPC::Matemàtiques i estadística
Descripción
Sumario:In recent years, due to the relation between cognitive control and mathematical concept of control dynamical systems, there has been growing interest in the descriptive analysis of complex networks with linear dynamics, permeating many aspects from everyday life, obtaining considerable advances in the description of their structural and dynamical properties. Nevertheless, much less effort has been devoted to studying the controllability of the dynamics taking place on them. Concretely, for complex systems is of interest to study the exact controllability, this measure is defined as the minimum set of controls that are needed to steer the whole system toward any desired state. In this paper, a revision of controllability concepts is presented and provides conditions for exact controllability for the multiagent systems