Online Learning a Symbolic Abstraction of Actions in Hierarchical RL with Formal Methods - Inria - Institut national de recherche en sciences et technologies du numérique
Communication Dans Un Congrès Année : 2024

Online Learning a Symbolic Abstraction of Actions in Hierarchical RL with Formal Methods

Résumé

Reasoning about actions, planning in open-ended learning requires a hierarchical abstractions of action models and benefits from the use of symbolic methods for goal representation as they structure knowledge for efficient and transferable learning. However, the existing Hierarchical Reinforcement Learning (HRL) approaches relying on symbolic reasoning are often limited as they require a manual goal representation. A symbolic goal representation must preserve information about the environment dynamics. We propose an automatic subgoal discovery via an emergent representation that abstracts (i.e., groups together) sets of environment states that have similar roles in the task, using formal verification methods. We create a HRL algorithm that learns online this representation along with the policies. We show on navigation tasks that the learned representation is interpretable and results in data efficiency.
Fichier principal
Vignette du fichier
Nguyen2024A.pdf (476.98 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04858759 , version 1 (29-12-2024)

Licence

Identifiants

  • HAL Id : hal-04858759 , version 1

Citer

Sao Mai Nguyen. Online Learning a Symbolic Abstraction of Actions in Hierarchical RL with Formal Methods. ActSynt@ECAI2024, Oct 2024, Santiago de Compostela, Spain. ⟨hal-04858759⟩
0 Consultations
0 Téléchargements

Partager

More