An Information-Theoretic Perspective on Intrinsic Motivation in Reinforcement Learning: A Survey
Entropy · 10.02.2023 · 25 (2) · 327
DOI: 10.3390/e25020327 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift Entropy, online erschienen am 10.02.2023. Verfasst von Arthur Aubret und 2 weiteren Autor:innen. Beteiligte Einrichtung: Univ Lyon, UCBL, CNRS, INSA Lyon, LIRIS, UMR5205, 69622 Villeurbanne, France. Die Arbeit wurde bisher 48-mal zitiert (Crossref) und nennt 190 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: ANR project DeLiCio.
Abstract (Originalsprache)
The reinforcement learning (RL) research area is very active, with an important number of new contributions, especially considering the emergent field of deep RL (DRL). However, a number of scientific and technical challenges still need to be resolved, among which we acknowledge the ability to abstract actions or the difficulty to explore the environment in sparse-reward settings which can be addressed by intrinsic motivation (IM). We propose to survey these research works through a new taxonomy based on information theory: we computationally revisit the notions of surprise, novelty, and skill-learning. This allows us to identify advantages and disadvantages of methods and exhibit current outlooks of research. Our analysis suggests that novelty and surprise can assist the building of a hierarchy of transferable skills which abstracts dynamics and makes the exploration process more robust.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Univ Lyon, UCBL, CNRS, INSA Lyon, LIRIS, UMR5205, 69622 Villeurbanne, France.
ORCID-Profile: Arthur Aubret
Förderung
- ANR project DeLiCio
Zitieren
APA
Aubret, A., Matignon, L., & Hassas, S. (2023). An Information-Theoretic Perspective on Intrinsic Motivation in Reinforcement Learning: A Survey. Entropy, 25(2), 327. https://doi.org/10.3390/e25020327
Vancouver
Aubret A, Matignon L, Hassas S. An Information-Theoretic Perspective on Intrinsic Motivation in Reinforcement Learning: A Survey. Entropy. 2023;25(2):327. doi:10.3390/e25020327
BibTeX
@article{Aubret2023,
author = {Aubret, Arthur and Matignon, Laetitia and Hassas, Salima},
title = {An Information-Theoretic Perspective on Intrinsic Motivation in Reinforcement Learning: A Survey},
journal = {Entropy},
year = {2023},
volume = {25},
number = {2},
pages = {327},
doi = {10.3390/e25020327}
}
EUVIMED