Learning classifier systems: a gentle introduction

Abstract

Learning Classifier Systems were introduced in the 1970s by John H. Holland as highly adaptive, cognitive systems. More than 40 years later, the introduction of Stewart W. Wilson's XCS, a highly engineered classifier system model, has transformed them into a state-of-the-art machine learning system. Learning classifier systems can effectively solve data-mining problems, reinforcement learning problems, and also cognitive, robotics control problems. In comparison to other, non-evolutionary machine learning techniques, their performance is competitive or superior, dependent on the setup and problem. Learning classifier systems can work both online and offline, they are extremely flexible, applicable to a larger range of problems, and are highly adaptive. Moreover, system knowledge can be easily extracted, visualized, or even used to focus the progressive search on particular interesting subspaces. This tutorial provides a gentle introduction to learning classifier systems and their general functionality. It then surveys the current theoretical understanding of the systems. Finally, we provide a suite of current successful LCS applications and discuss the most promising areas for future applications and research directions.