MTMA-DDPG: A Deep Deterministic Policy Gradient Reinforcement Learning for Multi-task Multi-agent Environments - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

MTMA-DDPG: A Deep Deterministic Policy Gradient Reinforcement Learning for Multi-task Multi-agent Environments

Karim Hamadeh
  • Fonction : Auteur
  • PersonId : 1318728
Julia El Zini
  • Fonction : Auteur
  • PersonId : 1318729
Joudi Hajar
  • Fonction : Auteur
  • PersonId : 1318730
Mariette Awad
  • Fonction : Auteur
  • PersonId : 1318731

Résumé

Beyond Multi-Task or Multi-Agent learning, we develop in this work a multi-agent reinforcement learning algorithm to handle a multi-task environments. Our proposed algorithm, Multi-Task Multi-Agent Deep Deterministic Policy gradient, (MTMA-DDPG) (Code available at https://gitlab.com/awadailab/mtmaddpg ), extends its single task counterpart by running multiple tasks on distributed nodes and communicating parameters via pre-determined coefficients across the nodes. Parameter sharing is modulated through temporal decay of the communication coefficients. Training across nodes is parallelized without any centralized controller for different tasks, which opens horizons for flexible leveraging and parallel processing to improve MA learning.Empirically, we design different MA particle environments, where tasks are similar or heterogeneous. We study the performance of MTMA-DDPG in terms of reward, convergence, variance, and communication overhead. We demonstrate the improvement of our algorithm over its single-task counterpart, as well as the importance of a versatile technique to take advantage of parallel computing resources.
Fichier sous embargo
Fichier sous embargo
0 6 24
Année Mois Jours
Avant la publication
mercredi 1 janvier 2025
Fichier sous embargo
mercredi 1 janvier 2025
Connectez-vous pour demander l'accès au fichier

Dates et versions

hal-04317184 , version 1 (01-12-2023)

Licence

Identifiants

Citer

Karim Hamadeh, Julia El Zini, Joudi Hajar, Mariette Awad. MTMA-DDPG: A Deep Deterministic Policy Gradient Reinforcement Learning for Multi-task Multi-agent Environments. 18th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Jun 2022, Hersonissos, Greece. pp.270-281, ⟨10.1007/978-3-031-08333-4_22⟩. ⟨hal-04317184⟩
24 Consultations
1 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More