Emergence of a Symbolic Goal Representation with an Intelligent Tutoring System based on Intrinsic Motivation - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Poster Communications Year : 2023

Emergence of a Symbolic Goal Representation with an Intelligent Tutoring System based on Intrinsic Motivation

Abstract

Goal representation affects the performance of Hierarchical Reinforcement Learning (HRL) algorithms by decomposing complex problems into easier subtasks. Recent studies show that representations that preserve temporally abstract environment dynamics are successful in solving difficult problems with theoretical guarantees for optimality. These methods however cannot scale to tasks where environment dynamics increase in complexity. On the other hand, other efforts have tried to use spatial abstraction to mitigate the previous issues. Their limitations include scalability to high dimensional environments and dependency on prior knowledge. In this work, we propose a novel three-layer HRL algorithm that introduces, at different levels of the hierarchy, both a spatial and a temporal goal abstraction. We provide a theoretical study of the regret bounds of the learned policies. We evaluate the approach on complex continuous control tasks, demonstrating the effectiveness of spatial and temporal abstractions learned by this approach.
Fichier principal
Vignette du fichier
main.pdf (1.83 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04403613 , version 1 (18-01-2024)

Licence

Attribution

Identifiers

  • HAL Id : hal-04403613 , version 1

Cite

Mehdi Zadem, Sergio Mover, Sao Mai Nguyen. Emergence of a Symbolic Goal Representation with an Intelligent Tutoring System based on Intrinsic Motivation. NeurIPS 2023 - IMOL Workshop "Intrinsically-Motivated and Open-Ended Learning", Dec 2023, New Orleans (Louisiana), United States. IEEE, pp.423-428. ⟨hal-04403613⟩
22 View
24 Download

Share

Gmail Facebook X LinkedIn More