The Successor Representation as a model of behavioural flexibility - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

The Successor Representation as a model of behavioural flexibility

Résumé

Accounting for behavioural capabilities and flexibilities experimentally observed in animals is a major issue in computational neurosciences. In order to design a comprehensive algorithmic framework for this purpose, the model-free and model-based reinforcement learning (RL) components are generally taken as reference either in isolation or in combination. In this article, we consider the RL Successor Representation (SR) approach as an alternative. We compare it to the standard model-free and modelbased models on three relevant experimental data-sets. These modelling experiments demonstrate that SR is able to account better for several behavioural flexibilities while being algorithmically simpler.
Fichier principal
Vignette du fichier
ducarouge-sigaud.pdf (1.08 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01576352 , version 1 (23-08-2017)

Identifiants

  • HAL Id : hal-01576352 , version 1

Citer

Alexis Ducarouge, Olivier Sigaud. The Successor Representation as a model of behavioural flexibility. Journées Francophones sur la Planification, la Décision et l'Apprentissage pour la conduite de systèmes (JFPDA 2017), Jul 2017, Caen, France. ⟨hal-01576352⟩
191 Consultations
234 Téléchargements

Partager

Gmail Facebook X LinkedIn More