Upper Confidence Trees with Short Term Partial Information

Olivier Teytaud; Sébastien Flory; Ferrante Neri

doi:10.1007/978-3-642-20525-5

Back

Conference proceeding

Upper Confidence Trees with Short Term Partial Information

Olivier Teytaud, Sébastien Flory and Ferrante Neri

Vol.6624, pp.153-162

Lecture Notes in Computer Science

EvoGames 2011

2011

DOI: https://doi.org/10.1007/978-3-642-20525-5

Abstract

Mathematics

Optimization and Control

We show some mathematical links between partially observable (PO) games in which information is regu larly revealed, and simultaneous actions games. Using this, we study the extension of Monte-Carlo Tr ee Search algorithms to PO games and to games with simultaneous actions. We apply the results to Urb an Rivals, a free PO internet card game with more than 10 millions of registered users.

Metrics

Details

Title: Upper Confidence Trees with Short Term Partial Information
Creators: Olivier Teytaud - Laboratoire de Recherche en Informatique
Sébastien Flory - BOOSTR
Ferrante Neri - Department of Electrical and Electronic Engineering
Publication Details: Vol.6624, pp.153-162
Conference: EvoGames 2011
Series: Lecture Notes in Computer Science
Publisher: Springer
Publication Date: 2011
Identifiers: 99783566002346
Academic Unit: School of Computer Science and Electronic Engineering
Language: English
Resource Type: Conference proceeding

Upper Confidence Trees with Short Term Partial Information

Abstract

Metrics

Details

Usage Policy