Evolution of reward functions for reinforcement learning

Created by W.Langdon from gp-bibliography.bib Revision:1.3973

@InProceedings{Niekum:2011:GECCOcomp,
  author =       "Scott Niekum and Lee Spector and Andrew Barto",
  title =        "Evolution of reward functions for reinforcement
                 learning",
  booktitle =    "GECCO '11: Proceedings of the 13th annual conference
                 companion on Genetic and evolutionary computation",
  year =         "2011",
  editor =       "Natalio Krasnogor and Pier Luca Lanzi and 
                 Andries Engelbrecht and David Pelta and Carlos Gershenson and 
                 Giovanni Squillero and Alex Freitas and 
                 Marylyn Ritchie and Mike Preuss and Christian Gagne and 
                 Yew Soon Ong and Guenther Raidl and Marcus Gallager and 
                 Jose Lozano and Carlos Coello-Coello and Dario Landa Silva and 
                 Nikolaus Hansen and Silja Meyer-Nieberg and 
                 Jim Smith and Gus Eiben and Ester Bernado-Mansilla and 
                 Will Browne and Lee Spector and Tina Yu and Jeff Clune and 
                 Greg Hornby and Man-Leung Wong and Pierre Collet and 
                 Steve Gustafson and Jean-Paul Watson and 
                 Moshe Sipper and Simon Poulding and Gabriela Ochoa and 
                 Marc Schoenauer and Carsten Witt and Anne Auger",
  isbn13 =       "978-1-4503-0690-4",
  keywords =     "genetic algorithms, genetic programming, Genetics
                 based machine learning: Poster",
  pages =        "177--178",
  month =        "12-16 " # jul,
  organisation = "SIGEVO",
  address =      "Dublin, Ireland",
  DOI =          "doi:10.1145/2001858.2001957",
  publisher =    "ACM",
  publisher_address = "New York, NY, USA",
  abstract =     "The reward functions that drive reinforcement learning
                 systems are generally derived directly from the
                 descriptions of the problems that the systems are being
                 used to solve. In some problem domains, however,
                 alternative reward functions may allow systems to learn
                 more quickly or more effectively. Here we describe work
                 on the use of genetic programming to find novel reward
                 functions that improve learning system performance. We
                 briefly present the core concepts of our approach, our
                 motivations in developing it, and reasons to believe
                 that the approach has promise for the production of
                 highly successful adaptive technologies. Experimental
                 results are presented and analysed in our full report
                 [3].",
  notes =        "Also known as \cite{2001957} Distributed on CD-ROM at
                 GECCO-2011.

                 ACM Order Number 910112.",
}

Genetic Programming entries for Scott Niekum Lee Spector Andrew G Barto

Citations