Neural Fitted Actor-Critic

Matthieu Zimmer 1, 2 Yann Boniface 1 Alain Dutech 2
1 CORTEX - Neuromimetic intelligence
Inria Nancy - Grand Est, LORIA - AIS - Department of Complex Systems, Artificial Intelligence & Robotics
2 MAIA - Autonomous intelligent machine
Inria Nancy - Grand Est, LORIA - AIS - Department of Complex Systems, Artificial Intelligence & Robotics
Abstract : A novel reinforcement learning algorithm that deals with both continuous state and action spaces is proposed. Domain knowledge requirements are kept minimal by using non-linear estimators and since the algorithm does not need prior trajectories or known goal states. The new actor-critic algorithm is on-policy, offline and model-free. It considers discrete time, stationary policies, and maximizes the discounted sum of rewards. Experimental results on two common environments, showing the good performance of the proposed algorithm, are presented.
Type de document :
Communication dans un congrès
European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2016), Apr 2016, Bruges, Belgium. ESANN 2016 proceedings, 〈https://www.elen.ucl.ac.be〉
Liste complète des métadonnées

Littérature citée [16 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01350651
Contributeur : Matthieu Zimmer <>
Soumis le : mardi 23 août 2016 - 17:27:35
Dernière modification le : dimanche 11 décembre 2016 - 17:41:41
Document(s) archivé(s) le : jeudi 24 novembre 2016 - 12:37:25

Fichier

main.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01350651, version 1

Citation

Matthieu Zimmer, Yann Boniface, Alain Dutech. Neural Fitted Actor-Critic. European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2016), Apr 2016, Bruges, Belgium. ESANN 2016 proceedings, 〈https://www.elen.ucl.ac.be〉. 〈hal-01350651〉

Partager

Métriques

Consultations de la notice

420

Téléchargements de fichiers

283