Bias-Optimal Incremental Learning of Control Sequences for Virtual Robots

Created by W.Langdon from gp-bibliography.bib Revision:1.3872

@InProceedings{schmidhuber:2004:IAS,
  author =       "Juergen Schmidhuber and Viktor P Zhumatiy and 
                 Matteo Gagliolo",
  title =        "Bias-Optimal Incremental Learning of Control Sequences
                 for Virtual Robots",
  booktitle =    "Procceedings of the eigth conference on Intelligent
                 Autonomous Systems, IAS-8",
  year =         "2004",
  pages =        "658--665",
  address =      "Amsterdam",
  keywords =     "genetic algorithms, genetic programming",
  URL =          "ftp://ftp.idsia.ch/pub/juergen/snakeias.pdf",
  abstract =     "Learning and planning control is hard. The search
                 space of traditional planners consists of sequences of
                 primitive actions. To exploit reusable subsequences and
                 other algorithmic regularities, however, we should
                 instead search the general space of programs that
                 compute action sequences. Such programs may invoke very
                 fast thinking actions consuming only nanoseconds (such
                 as conditional jumps to certain code addresses) as well
                 as very slow control actions consuming seconds in the
                 real world (such as
                 stretch-arm-until-obstacle-sensation). What is an
                 optimal way of allocating time to tests of such
                 non-homogeneous programs? What is an optimal way of
                 reusing experience with previous tasks to learn
                 solutions to new tasks? One answer is given by the
                 recent Optimal Ordered Problem Solver OOPS, a
                 near-bias-optimal incremental extension of Levin's
                 nonincremental universal search, which we apply to
                 virtual robotics for the first time: our snake robot
                 uses OOPS to learn to walk and jump in a partially
                 observable environment (POMDP) with a huge state/action
                 space.",
}

Genetic Programming entries for Jurgen Schmidhuber Viktor P Zhumatiy Matteo Gagliolo

Citations