Perceiving Agent Collaborative Sonic Exploration In Interactive Reinforcement Learning

Hugo Scurto 1 Frédéric Bevilacqua 1 Baptiste Caramiaux 2, 1
1 Equipe Interaction Son Musique Mouvement
STMS - Sciences et Technologies de la Musique et du Son : UMR 9912
2 EX-SITU - Extreme Interaction
LRI - Laboratoire de Recherche en Informatique, Inria Saclay - Ile de France
Abstract : We present the first implementation of a new framework for sound and music computing, which allows humans to explore musical environments by communicating feedback to an artificial agent. It is based on an interactive reinforcement learning workflow, which enables agents to incrementally learn how to act on an environment by balancing exploitation of human feedback knowledge and exploration of new musical content. In a controlled experiment , participants successfully interacted with these agents to reach a sonic goal in two cases of different complexities. Subjective evaluations suggest that the exploration path taken by agents, rather than the fact of reaching a goal, may be critical to how agents are perceived as collaborative. We discuss such quantitative and qualitative results and identify future research directions toward deploying our "co-exploration" approach in real-world contexts.
Type de document :
Communication dans un congrès
Proceedings of the 15th Sound and Music Computing Conference (SMC 2018), Jul 2018, Limassol, Cyprus. 2018
Liste complète des métadonnées

Littérature citée [6 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01849074
Contributeur : Hugo Scurto <>
Soumis le : mercredi 25 juillet 2018 - 14:44:24
Dernière modification le : jeudi 22 novembre 2018 - 14:10:33
Document(s) archivé(s) le : vendredi 26 octobre 2018 - 15:06:40

Fichier

smc2018_paper_153_camera_ready...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01849074, version 1

Citation

Hugo Scurto, Frédéric Bevilacqua, Baptiste Caramiaux. Perceiving Agent Collaborative Sonic Exploration In Interactive Reinforcement Learning. Proceedings of the 15th Sound and Music Computing Conference (SMC 2018), Jul 2018, Limassol, Cyprus. 2018. 〈hal-01849074〉

Partager

Métriques

Consultations de la notice

264

Téléchargements de fichiers

63