Title :
What should a classifier system learn?
Author_Institution :
Sch. of Comput. Sci., Birmingham Univ., UK
Abstract :
We consider the issue of how a classifier system should learn to represent a Boolean function. We identify four properties which may be desirable of a representation; that it be complete, accurate, minimal and nonoverlapping, and distinguish variations on two of these properties for the XCS system. We question whether the bias against overlapping rules evident in some systems is appropriate, and find that XCS´s bias against overlapping rules is very strong
Keywords :
Boolean functions; learning systems; pattern classification; Boolean function; Learning Classifier Systems; XCS system; classifier system; overlapping rules; Boolean functions; Computer science; Machine learning; Monitoring; Statistics; Testing;
Conference_Titel :
Evolutionary Computation, 2001. Proceedings of the 2001 Congress on
Conference_Location :
Seoul
Print_ISBN :
0-7803-6657-3
DOI :
10.1109/CEC.2001.934268