Thesis Topic Details

Topic ID:
3553
Title:
Generating Game-Specific Knowledge to Improve Monte Carlo Tree Search
Supervisor:
Michael Thielscher
Research Area:
Artificial Intelligence, Machine Learning
Associated Staff
Assessor:
Alan Blair
Topic Details
Status:
Active
Type:
R & D
Programs:
CS CE BIOM BINF SE
Group Suitable:
No
Industrial:
No
Pre-requisites:
COMP3411 Artificial Intelligence or COMP9417 Machine Learning
Description:
General game players (GGPs) are computer systems able to play strategy games based solely on formal game descriptions supplied at "runtime" (in other words, they don't know the rules until the game starts). Unlike specialised game players, such as Deep Blue, general game players cannot rely on algorithms designed in advance for specific games; they must discover such algorithms themselves.

One of the most successful approaches to general game playing is the use of uninformed search in the form of Monte Carlo simulations to determine the best moves. But expertise of a GGP depends on intelligence on the part of the game player and not just intelligence of the programmer of the game player. This project therefore aims at developing and implementing techniques to automatically generate additional knowledge about a game from the mere rules in order to improve the quality of Monte Carlo Tree Search. The project involves implementing and testing these methods with an existing GGP developed at our School.

If successful, this research could be published in a high-profile conference or journal.
Comments:
M. Genesereth, M. Thielscher. General Game Playing. Synthesis Lectures on Artificial Intelligence and Machine Learning 2014.
http://www.morganclaypool.com/doi/abs/10.2200/S00564ED1V01Y201311AIM024

M. Genesereth, N. Love, B. Pell: General Game Playing: Overview of the AAAI Competition, AAAI Magazine, Spring 2005.
http://logic.stanford.edu/classes/cs227/2014/readings/aaai.pdf
Past Student Reports
 
No Reports Available. Contact the supervisor for more information.

Check out all available reports in the CSE Thesis Report Library.

NOTE: only current CSE students can login to view and select reports to download.