Abstract
In this paper, the active learning mechanism is proposed to be used in classifier systems to cope with complex problems: an, intelligent agent leaves its own signals in the environment and later collects and employs them to assist its learning process. Principles and components of the mechanism are outlined, followed by the introduction of its preliminary implementation in an actual system. An experiment with the system in a dynamic problem is then introduced, together with discussions over its results. The paper is concluded by pointing out some possible improvements that can be made to the proposed framework.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Holland J H, Reitman J S. Cognitive Systems Based on Adaptive Algorithms. Pattern Directed Inference Systems, Academic Press, NY, 1978, pp.313–329.
Wilson S W. ZCS: A zeroth level classifier system.Evolutionary Computation, 1994, 2(1): 1–18.
Wilson S W. Classifier systems and the Animat problem.Machine Learning, 1987, 2(3): 199–228.
Booker L B. Intelligent behavior as an adaptation to the task environment.Dissertations Abstracts International, 1982, 43(2), 469B.
Wilson S W. Adaptive “cortical” pattern recognition. InProc. of the 1st International Conference on Genetic Algorithm and Their Applications (ICGA'85), 1985, pp.188–196.
Holland J H. Escaping brittleness: The possibilities of general-purpose learning algorithms applied to parallel rule-based systems. Machine Learning, An Artificial Intelligence Approach, Vol.2, Morgan Kaufmann, Los Altos, CA. 1986.
Goldberg D E. Genetic Algorithms in Search, Optimization & Machine Learning. Addison-Wesley, Reading, MA, Jan. 1989.
Goldberg G E. Genetic algorithms and rule learning in dynamic system control. InProc. 1st International Conference on Genetic Algorithm and Their Applications (ICGA'85), 1985, pp.8–15.
Xi Haifeng, Luo Yupin, Jiang Rui, Hu Dongcheng. Incorporation of signal store into classifier systems: Principle, and experiments. InProc. 1997 IEEE International Conference on Intelligent Processing Systems (ICIPS'97), 1997, pp.905–909.
Author information
Authors and Affiliations
Additional information
This work was supported by the doctor training program foundation of the State Education Commission of China.
XI Haifeng was born in 1972. He received his BS and MS degrees from Tsinghua University in July 1995 and June 1997 respectively. His main research interests are distributed systems, artificial intelligence, GAs and their applications.
LUO Yupin was born in 1959. He graduated from Hunan University in 1982. He received his M.S. and Ph.D. degrees from Nagoya Institute of Technology, Japan in March 1987 and March 1990 respectively. He is now an Associate Professor of Department of Automation, Tsinghua University. His current research interests are communications network-related graph theory, distributed computing, group intelligence systems, AI and application of GAs.
YANG Shiyuan was born in Shanghai, China, in Nov. 1945. He received his B.S. and M.S. degrees from Tsinghua University in 1970 and 1981 respectively. From 1989 to 1995, he was an Associate Professor of Department of Automation, Tsinghua University. Since July 1995, he has been a Professor of the same department. He is the Co-Chairman of the Tolerant Computing Committee of China, senior member of Chinese Electronic Institute and senior member of IEEE. His current research interests are fault detection and diagnosis in digital and analog circuits, AI and its application, application of ANN to fault diagnosis, reliability design of control system and EMC of electronic equipment.
Rights and permissions
About this article
Cite this article
Xi, H., Luo, Y. & Yang, S. An approach to active learning for classifier systems. J. Comput. Sci. & Technol. 14, 372–378 (1999). https://doi.org/10.1007/BF02948739
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02948739