Papers available: modular genetic programming

[rosca@cs.rochester.edu]

The following papers on discovery of subroutines in Genetic
Programming are now available for retrieval via ftp.

Ftp: ftp://ftp.cs.rochester.edu/pub/u/rosca/gp
WWW: http://www.cs.rochester.edu/u/rosca/research.html

Comments and suggestions are welcome.

Justinian --

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Justinian Rosca                    Internet: rosca at cs.rochester.edu
University of Rochester            Office: (716) 275-1174
Department of Computer Science     Fax: (716) 461-2018
Rochester, NY 14627-0226           WWW: http://www.cs.rochester.edu/u/rosca/



             EVOLUTION-BASED DISCOVERY OF HIERARCHICAL BEHAVIORS
                      J.P. Rosca and D.H. Ballard  
                      AAAI-96, The MIT Press, 1996

       ftp://ftp.cs.rochester.edu/pub/u/rosca/gp/96.aaai.ps.gz 
                       (7 pages; 100k compressed)
Abstract:   

  The complexity of policy learning in a reinforcement
  learning task deteriorates primarily with the increase of the number
  of observations.  Unfortunately, the number of observations may be
  unacceptably high even for simple problems. In order to cope with
  the scale up problem we adopt procedural representations of
  policies. Procedural representations have two advantages.  First
  they are implicit, allowing for good inductive generalization over a
  very large set of input states. Second they facilitate
  modularization. In this paper we compare several randomized
  algorithms for learning modular procedural representations. The main
  algorithm, called Adaptive Representation through Learning (ARL) is
  a genetic programming extension that relies on the discovery of
  subroutines.  ARL is suitable for learning hierarchies of
  subroutines and for constructing policies to complex tasks. When the
  learning problem cannot be solved because the specification is too
  loose and the domain is not well understood, ARL will discover
  regularities in the problem environment in the form of subroutines,
  which often lead to an easier problem solving. ARL was successfully
  tested on a typical reinforcement learning problem of controlling an
  agent in a dynamic and non-deterministic environment where the
  discovered subroutines correspond to agent behaviors.


             DISCOVERY OF SUBROUTINES IN GENETIC PROGRAMMING
                     J.P. Rosca and D.H. Ballard
               In: Advances in Genetic Programming II
        Edited by P. Angeline and K. Kinnear Jr.  MIT Press, 1996.

       ftp://ftp.cs.rochester.edu/pub/u/rosca/gp/96.aigp2.dsgp.ps.gz 
                      (25 pages; 439k compressed)
Abstract:   

  A fundamental problem in learning from observation and interaction
  with an environment is defining a good representation, that is a
  representation which captures the underlying structure and
  functionality of the domain.  This chapter discusses an extension of
  the genetic programming (GP) paradigm based on the idea that
  subroutines obtained from blocks of good representations act as
  building blocks and may enable a faster evolution of even better
  representations.  This GP extension algorithm is called adaptive
  representation through learning (ARL). It has built-in mechanisms
  for (1) creation of new subroutines through discovery and
  generalization of blocks of code; (2) deletion of subroutines. The
  set of evolved subroutines extracts common knowledge emerging during
  the evolutionary process and acquires the necessary structure for
  solving the problem. ARL was successfully tested on the problem of
  controlling an agent in a dynamic and non-deterministic environment.
  Results with the automatic discovery of subroutines show the
  potential to better scale up the GP technique to complex problems.