Abstract
In machine learning “forgetting” little used or redundant information can be seen as a sensible strategy directed at the overall management of specific and limited computational resources. This paper describes new learning rules for the ART2 neural network model of category learning that facilitates forgetting without additional node features or subsystems and which preserves the main characteristics of the classic ART2 model. We consider that this approach is straightforward and is arguably biological plausible. The new learning rules drop the specification within the classic ART2 model that learning should only occur at the winning node. Classic ART2 learning rules are presented as a particular case of these new rules. The model increases system adaptability to continually changing or complex input domains. This allows the system to maintain information in a manner which is consistent with its use and allows system resources to be dynamically allocated in a way that is consistent with observations made of biological learning.
Preview
Unable to display preview. Download preview PDF.
References
Baddley, A.: The Psychology of Memory. New York Basis Books (1976)
Carpenter G., Grossberg S.: ART2: Self-Organization of Stable Category Recognition Codes for Analog Input Patterns. Applied Optics 26 (1987)4916–4930
Corsini R.: Encyclopaedia of Psychology. John Wiley & Sons, vol. 2 (1984)
Ebbinghaus H.: Memory: A Contribution to Experimental Psychology. New York Dover (1964)
Fritzke B.: Unsupervised Clustering with Growing Cell Structures. Proc. of the IJCNN'91 Seattle (IEEE) (1991)
Fritzke B.: Let It Grow — Self-Organizing Feature Map with Problem Dependent Cell Structure. Proc. of the ICANN'91 Helsinki (1991)
Grossberg S.: Adaptive Pattern Classification and Universal Recoding. II: Feedback, Expectation, Olfaction and Illusion. Biol. Cybern. 23 (1976) 187
Grossberg S.: How Does a Brain Build a Cognitive Code? Psychological Review. 1 (1980) 1–51
Hebb D.: The Organization and Behaviour. New York: Witey (1949)50
Keuchel H.: Putcamer E., Zimmer U.: Learning and Forgetting Surface Classification with Dynamic Neural Networks. Proc. of the ICANN'93. Amsterdam IX (1993)
Kohonen T.: Statistical Pattern Recognition Revisited. Advanced Network Computers R. Eckmiller (ed.) (1990)
Kohonen T.: Self-Organization and Associative Memory. Springer Verlag Berlin (1984)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nachev, A., Griffith, N., Gerov, A. (1998). Dynamic learning — An approach to forgetting in ART2 neural networks. In: Giunchiglia, F. (eds) Artificial Intelligence: Methodology, Systems, and Applications. AIMSA 1998. Lecture Notes in Computer Science, vol 1480. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0057458
Download citation
DOI: https://doi.org/10.1007/BFb0057458
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64993-9
Online ISBN: 978-3-540-49793-6
eBook Packages: Springer Book Archive