Discovering and Exploiting Sparse Rewards in a Learned Behavior Space - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Evolutionary Computation Année : 2023

Discovering and Exploiting Sparse Rewards in a Learned Behavior Space

Giuseppe Paolo
Miranda Coninx
  • Fonction : Auteur
Alban Laflaquière
  • Fonction : Auteur
  • PersonId : 1096132

Résumé

Learning optimal policies in sparse rewards settings is difficult as the learning agent has little to no feedback on the quality of its actions. In these situations, a good strategy is to focus on exploration, hopefully leading to the discovery of a reward signal to improve on. A learning algorithm capable of dealing with this kind of settings has to be able to (1) explore possible agent behaviors and (2) exploit any possible discovered reward. Efficient exploration algorithms have been proposed that require to define a behavior space, that associates to an agent its resulting behavior in a space that is known to be worth exploring. The need to define this space is a limitation of these algorithms. In this work, we introduce STAX, an algorithm designed to learn a behavior space on-the-fly and to explore it while efficiently optimizing any reward discovered. It does so by separating the exploration and learning of the behavior space from the exploitation of the reward through an alternating two-steps process. In the first step, STAX builds a repertoire of diverse policies while learning a low-dimensional representation of the high-dimensional observations generated during the policies evaluation. In the exploitation step, emitters are used to optimize the performance of the discovered rewarding solutions. Experiments conducted on three different sparse reward environments show that STAX performs comparably to existing baselines while requiring much less prior information about the task as it autonomously builds the behavior space.
Fichier principal
Vignette du fichier
2111.01919.pdf (2.73 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04239313 , version 1 (12-10-2023)

Identifiants

Citer

Giuseppe Paolo, Miranda Coninx, Alban Laflaquière, Stéphane Doncieux. Discovering and Exploiting Sparse Rewards in a Learned Behavior Space. Evolutionary Computation, inPress, pp.1-28. ⟨hal-04239313⟩
26 Consultations
18 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More