Solving a Goal-planning task in the MASH project

Jean-Baptiste Hoock 1, 2 Jacques Bibai 2, 3
2 TAO - Machine Learning and Optimisation
LRI - Laboratoire de Recherche en Informatique, UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France, CNRS - Centre National de la Recherche Scientifique : UMR8623
Abstract : The MASH project is a collaborative platform with the aim to experiment different methods in an unknown environment of large size. The application is a goal-planning task in a 3D video game where runs are expensive. Moreover, there is no prior knowledge, the decisions have unknown semantics, observations on the environment are partial and of big size and accomplishing the task by taking random decisions always requires a very long run. So, solving this task is a big challenge. In this paper, we extend Monte-Carlo Tree Search, which has been proved very effective for applications in which simulating is easy and fast, to contexts in which there are only "real" expensive runs. This generic approach combines Clustering and Monte-Carlo Tree Search.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [8 references]  Display  Hide  Download

https://hal.inria.fr/hal-00738073
Contributor : Jean-Baptiste Hoock <>
Submitted on : Wednesday, October 24, 2012 - 2:02:14 PM
Last modification on : Thursday, April 5, 2018 - 12:30:12 PM
Document(s) archivé(s) le : Friday, January 25, 2013 - 3:41:25 AM

File

clustering.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00738073, version 1

Collections

Citation

Jean-Baptiste Hoock, Jacques Bibai. Solving a Goal-planning task in the MASH project. The 2012 Conference on Technologies and Applications of Artificial Intelligence (TAAI 2012), Nov 2012, Tainan, Taiwan. ⟨hal-00738073⟩

Share

Metrics

Record views

418

Files downloads

236