Stabilized Nested Rollout Policy Adaptation

Cazenave, Tristan; Sevestre, Jean-Baptiste; Toulemont, Matthieu

Cazenave, Tristan; Sevestre, Jean-Baptiste; Toulemont, Matthieu (2020), Stabilized Nested Rollout Policy Adaptation, Monte Carlo Search 2020, IJCAI Workshop, 2021-01, Yokohama (virtual), Japan

View/Open

StabilizedNRPA.pdf (282.3Kb)

Type

Communication / Conférence

Date

2020

Conference title

Monte Carlo Search 2020, IJCAI Workshop

Conference date

2021-01

Conference city

Yokohama (virtual)

Conference country

Japan

Metadata

Show full item record

Author(s)

Cazenave, Tristan
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Sevestre, Jean-Baptiste
Toulemont, Matthieu

Abstract (EN)

Nested Rollout Policy Adaptation (NRPA) is a Monte Carlo search algorithm for single player games. In this paper we propose to modify NRPA in order to improve the stability of the algorithm. Experiments show it improves the algorithm for different application domains: SameGame, Traveling Salesman with Time Windows and Expression Discovery.

Subjects / Keywords

Monte Carlo Tree Search