Mean Field Games Flock! The Reinforcement Learning Way - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès IJCAI Année : 2021

Mean Field Games Flock! The Reinforcement Learning Way

Sarah Perrin
  • Fonction : Auteur
Mathieu Laurière
  • Fonction : Auteur
Julien Pérolat
  • Fonction : Auteur
Matthieu Geist
  • Fonction : Auteur
  • PersonId : 790158
  • IdRef : 142341819
Romuald Élie
  • Fonction : Auteur
Olivier Pietquin
  • Fonction : Auteur

Résumé

We present a method enabling a large number of agents to learn how to flock, which is a natural behavior observed in large populations of animals. This problem has drawn a lot of interest but requires many structural assumptions and is tractable only in small dimensions. We phrase this problem as a Mean Field Game (MFG), where each individual chooses its acceleration depending on the population behavior. Combining Deep Reinforcement Learning (RL) and Normalizing Flows (NF), we obtain a tractable solution requiring only very weak assumptions. Our algorithm finds a Nash Equilibrium and the agents adapt their velocity to match the neighboring flock's average one. We use Fictitious Play and alternate: (1) computing an approximate best response with Deep RL, and (2) estimating the next population distribution with NF. We show numerically that our algorithm learn multi-group or high-dimensional flocking with obstacles.

Dates et versions

hal-03416242 , version 1 (05-11-2021)

Identifiants

Citer

Sarah Perrin, Mathieu Laurière, Julien Pérolat, Matthieu Geist, Romuald Élie, et al.. Mean Field Games Flock! The Reinforcement Learning Way. IJCAI, Aug 2021, Montreal, Canada. ⟨hal-03416242⟩
41 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More