Evolutionary Intelligence

Learning Classifier Systems are a genetic-based machine learning paradigm introduced by
John Holland in 1976. They are rule-based systems
which allow different representations, rule discovery mechanisms, and credit assignment
schemes. Current applications of Learning
Classifier Systems range from data mining to on-line cognitive control.
Learning Classifier Systems are a very active research field with a dedicated workshop
(the International Workshop on Learning Classifier Systems - IWLCS) organized every
year since 1999.
Recently, the success in parallelization by the IBM Watson team and application of LCSs to
data mining problems are receiving a great deal of attention. In addition, the
field also benefits from advances in reinforcement learning, adaptive filtering, and machine
learning that are continuously exploited to improve and extend LCSs.



Background Material:
Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to
Biology, Control, and Artificial Intelligence (Complex Adaptive Systems) [Kindle Edition]
by John H. Holland


Next