Goal-Conditioned Reinforcement Learning with Imagined Subgoals - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Goal-Conditioned Reinforcement Learning with Imagined Subgoals

Résumé

Goal-conditioned reinforcement learning endows an agent with a large variety of skills, but it often struggles to solve tasks that require more temporally extended reasoning. In this work, we propose to incorporate imagined subgoals into policy learning to facilitate learning of complex tasks. Imagined subgoals are predicted by a separate high-level policy, which is trained simultaneously with the policy and its critic. This high-level policy predicts intermediate states halfway to the goal using the value function as a reachability metric. We don't require the policy to reach these subgoals explicitly. Instead, we use them to define a prior policy, and incorporate this prior into a KL-constrained policy iteration scheme to speed up and regularize learning. Imagined subgoals are used during policy learning, but not during test time, where we only apply the learned policy. We evaluate our approach on complex robotic navigation and manipulation tasks and show that it outperforms existing methods by a large margin.
Fichier principal
Vignette du fichier
2107.00541.pdf (1.4 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03470313 , version 1 (08-12-2021)

Identifiants

Citer

Elliot Chane-Sane, Cordelia Schmid, Ivan Laptev. Goal-Conditioned Reinforcement Learning with Imagined Subgoals. ICML 2021 - Thirty-eighth International Conference on Machine Learning, Jul 2021, Virtual, France. ⟨hal-03470313⟩
57 Consultations
69 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More