Dynamic Programming in Distributional Reinforcement Learning - Archive ouverte HAL Accéder directement au contenu
Rapport Année : 2020

Dynamic Programming in Distributional Reinforcement Learning

Elie Odin
  • Fonction : Auteur
  • PersonId : 1093451

Résumé

The classic approach to reinforcement learning is limited in that it only predicts the expected return. The specialized literature has long tried to remedy this problem by studying risk-sensitive models, but the distributional approach will not emerge until 2017. Since the seminal article M. G. Bellemare, Dabney, and Munos 2017 and the state-of-the-art performance of the C51 algorithm in the ATARI 2600 suite of benchmark tasks (M. G. Bellemare, Naddaf, et al. 2013), research has focused on understanding the behaviour of distributional algorithms. In this paper we place Bellemare's original results in distributional dynamic programming in parallel with the classic results.
Fichier principal
Vignette du fichier
Dynamic Programming in Distributional Reinforcement Learning.pdf (639.98 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03168889 , version 1 (14-03-2021)

Identifiants

  • HAL Id : hal-03168889 , version 1

Citer

Elie Odin, Arthur Charpentier. Dynamic Programming in Distributional Reinforcement Learning. [Research Report] Université du Québec à Montréal. 2020. ⟨hal-03168889⟩

Collections

LARA
405 Consultations
363 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More