Patrick F. Riley's Publications

Sorted by DateClassified by Publication TypeClassified by Research Category

Advice Generation from Observed Execution: Abstract Markov Decision Process Learning

Patrick Riley and Manuela Veloso. Advice Generation from Observed Execution: Abstract Markov Decision Process Learning. In Proceedings of the Nineteenth National Conference on Artificial Intelligence (AAAI-2004), 2004.

Download

[PDF]129.0kB  

Abstract

An advising agent, a coach, provides advice to other agents about how to act. In this paper we contribute an advice generation method using observations of agents acting in an environment. Given an abstract state definition and partially specified abstract actions, the algorithm extracts a Markov Chain, infers a Markov Decision Process, and then solves the MDP (given an arbitrary reward signal) to generate advice. We evaluate our work in a simulated robot soccer environment and experimental results show improved agent performance when using the advice generated from the MDP for both a sub-task and the full soccer game.

Additional Information

Online Appendix

BibTeX

@InProceedings{AAAI04-abstract-MDP,
  author =	 {Patrick Riley and Manuela Veloso},
  title =	 {Advice Generation from Observed Execution: Abstract
                  {M}arkov Decision Process Learning},
  year =	 2004,
  booktitle =	 aaai2004,
  abstract =	 { An advising agent, a coach, provides advice to
                  other agents about how to act. In this paper we
                  contribute an advice generation method using
                  observations of agents acting in an
                  environment. Given an abstract state definition and
                  partially specified abstract actions, the algorithm
                  extracts a Markov Chain, infers a Markov Decision
                  Process, and then solves the MDP (given an arbitrary
                  reward signal) to generate advice. We evaluate our
                  work in a simulated robot soccer environment and
                  experimental results show improved agent performance
                  when using the advice generated from the MDP for
                  both a sub-task and the full soccer game. },
  bib2html_pubtype ={Refereed Conference},
  bib2html_rescat ={Coaching,Opponent and Teammate Modeling},
  bib2html_funding ={ActiveTemplates,RADAR},
  bib2html_extra_info={<a href="http://www.cs.cmu.edu/~pfr/appedices/2004aaai.html">Online Appendix</a>}
}

Generated by bib2html.pl (written by Patrick Riley ) on Thu Mar 31, 2005 16:21:00