In this article, we explore the use of genetic algorithms (GAs) as a k
ey element in the design and implementation of robust concept learning
systems. We describe and evaluate a GA-based system called GABIL that
continually learns and refines concept classification rules from its
interaction with the environment. The use of GAs is motivated by recen
t studies showing the effects of various forms of bias built into diff
erent concept learning systems, resulting in systems that perform well
on certain concept classes (generally, those well matched to the bias
es) and poorly on others. By incorporating a GA as the underlying adap
tive search mechanism, we are able to construct a concept learning sys
tem that has a simple, unified architecture with several important fea
tures. First, the system is surprisingly robust even with minimal bias
. Second, the system can be easily extended to incorporate traditional
forms of bias found in other concept learning systems. Finally, the a
rchitecture of the system encourages explicit representation of such b
iases and, as a result, provides for an important additional feature:
the ability to dynamically adjust system bias. The viability of this a
pproach is illustrated by comparing the performance of GABIL with that
of four other more traditional concept learners (AQ14, C4.5, ID5R, an
d IACL) on a variety of target concepts. We conclude with some observa
tions about the merits of this approach and about possible extensions.