Peter Stone's Selected Publications

Classified by TopicClassified by Publication TypeSorted by DateSorted by First Author Last NameClassified by Funding Source


HyperNEAT-GGP: A HyperNEAT-based Atari General Game Player

Matthew Hausknecht, Piyush Khandelwal, Risto Miikkulainen, and Peter Stone. HyperNEAT-GGP: A HyperNEAT-based Atari General Game Player. In Genetic and Evolutionary Computation Conference (GECCO), July 2012.

Download

[PDF]880.9kB  [postscript]3.3MB  

Abstract

This paper considers the challenge of enabling agents tolearn with as little domain-specific knowledge as possible. The maincontribution is HyperNEAT-GGP, a HyperNEAT-based General Game Playingapproach to Atari games. By leveraging the geometric regularitiespresent in the Atari game screen, HyperNEAT effectively evolvespolicies for playing two different Atari games, Asterix and Freeway.Results show that HyperNEAT-GGP outperforms existing benchmarks onthese games. HyperNEAT-GGP represents a step towards the ambitiousgoal of creating an agent capable of learning and seamlesslytransitioning between many different tasks.

BibTeX Entry

@InProceedings{GECCO12-Hausknecht,
title={HyperNEAT-GGP: A HyperNEAT-based Atari General Game Player},
author={Matthew Hausknecht and Piyush Khandelwal and Risto Miikkulainen and Peter Stone},
booktitle={Genetic and Evolutionary Computation Conference (GECCO)},
location={Philadelphia, Pennsylvania, USA},
month={July},
url="http://www.cs.utexas.edu/users/ai-lab/pub-view.php?PubID=127155",
year={2012},
abstract={This paper considers the challenge of enabling agents to
learn with as little domain-specific knowledge as possible. The main
contribution is HyperNEAT-GGP, a HyperNEAT-based General Game Playing
approach to Atari games. By leveraging the geometric regularities
present in the Atari game screen, HyperNEAT effectively evolves
policies for playing two different Atari games, Asterix and Freeway.
Results show that HyperNEAT-GGP outperforms existing benchmarks on
these games. HyperNEAT-GGP represents a step towards the ambitious
goal of creating an agent capable of learning and seamlessly
transitioning between many different tasks.},
}

Generated by bib2html.pl (written by Patrick Riley ) on Wed Jul 09, 2014 11:54:42