MESSI: Maximum Entropy Semi-Supervised Inverse Reinforcement Learning - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

MESSI: Maximum Entropy Semi-Supervised Inverse Reinforcement Learning

Michal Valko
Alessandro Lazaric
Mohammad Ghavamzadeh
  • Fonction : Auteur
  • PersonId : 868946

Résumé

A popular approach to apprenticeship learning (AL) is to formulate it as an inverse reinforcement learning (IRL) problem. The MaxEnt-IRL algorithm successfully integrates the maximum entropy principle into IRL and unlike its predecessors, it resolves the ambiguity arising from the fact that a possibly large number of policies could match the expert's behavior. In this paper, we study an AL setting in which in addition to the expert's trajectories, a number of unsupervised trajectories is available. We introduce MESSI, a novel algorithm that combines MaxEnt-IRL with principles coming from semi-supervised learning. In particular, MESSI integrates the unsupervised data into the MaxEnt-IRL framework using a pairwise penalty on trajectories. Empirical results in a highway driving and grid-world problems indicate that MESSI is able to take advantage of the unsupervised trajectories and improve the performance of MaxEnt-IRL.
Fichier principal
Vignette du fichier
audiffren2014messi.pdf (332.11 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01177446 , version 1 (16-07-2015)

Identifiants

  • HAL Id : hal-01177446 , version 1

Citer

Julien Audiffren, Michal Valko, Alessandro Lazaric, Mohammad Ghavamzadeh. MESSI: Maximum Entropy Semi-Supervised Inverse Reinforcement Learning. NIPS Workshop on Novel Trends and Applications in Reinforcement Learning, 2014, Montreal, Canada. ⟨hal-01177446⟩
239 Consultations
161 Téléchargements

Partager

Gmail Facebook X LinkedIn More