Foundations of Learning Classifier Systems

• Συγγραφή απο Οργανισμό/Αρχή: SpringerLink (Online service)
• Άλλοι συγγραφείς: Bull, Larry (Επιμελητής έκδοσης), Kovacs, Tim (Επιμελητής έκδοσης)
• Μορφή: Ηλεκτρονική πηγή Ηλ. βιβλίο
• Γλώσσα: English
• Έκδοση: Berlin, Heidelberg : Springer Berlin Heidelberg, 2005.
• Σειρά: Studies in Fuzziness and Soft Computing, 183
• Θέματα: Engineering.; Artificial intelligence.; Bioinformatics.; Applied mathematics.; Engineering mathematics.; Appl.Mathematics/Computational Methods of Engineering.; Artificial Intelligence (incl. Robotics).; Applications of Mathematics.

 This volume brings together recent theoretical work in Learning Classifier Systems (LCS), which is a Machine Learning technique combining Genetic Algorithms and Reinforcement Learning. It includes self-contained background chapters on related fields (reinforcement learning and evolutionary computation) tailored for a classifier systems audience and written by acknowledged authorities in their area - as well as a relevant historical original work by John Holland. 

 Section 1 – Rule Discovery. Population Dynamics of Genetic Algorithms. Approximating Value Functions in Classifier Systems. Two Simple Learning Classifier Systems. Computational Complexity of the XCS Classifier System. An Analysis of Continuous-Valued Representations for Learning Classifier Systems -- Section 2 – Credit Assignment. Reinforcement Learning: a Brief Overview. A Mathematical Framework for Studying Learning Classifier Systems. Rule Fitness and Pathology in Learning Classifier Systems. Learning Classifier Systems: A Reinforcement Learning Perspective. Learning Classifier Systems with Convergence and Generalization -- Section 3 – Problem Characterization. On the Classification of Maze Problems. What Makes a Problem Hard?