Publication: Learning sequences of rules using classifier systems with tags
Loading...
Identifiers
Publication date
1999-10
Defense date
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
IEEE
Impact
Export
Abstract
The objective of this paper was to obtain an encoding structure that would allow the genetic evolution of rules in such a manner that the number of rules and relationship in a classifier system (CS) would be learnt in the evolution process. For this purpose, an area that allows the definition of rule groups has been entered into the condition and message part of the encoded rules. This area is called internal tag. This term was coined because the system has some similarities with natural processes that take place in certain animal species, where the existence of tags allows them to communicate and recognize each other. Such CS is called a tag classifier system (TCS). The TCS has been tested in the game of draughts and compared with the classical CS. The results show an improving of the CS performance.
Description
IEEE International Conference on Systems, Man, and Cybernetics. Tokyo, 12-15 October 1999.
Keywords
Bibliographic citation
IEEE International Conference on Systems, Man, and Cybernetics, 1999 : IEEE SMC '99 Conference Proceedings. vol 5, p. 624-629