A New Geometry with Cross-Coupling of ART Networks
This paper demonstrates a new geometrical arrangement of adaptive resonance theory based network. Using method of minimal anatomies a neural network was constructed in an attempt to compare patterns. The anatomy incorporates two sub-networks coupled by feedback signals and an additional motor layer whose outputs reflect relationship or non-relationship among the compared patterns. Simulation results illustrates the network behaviors as emergent properties. The network with unsupervised learning is capable of generating self-defining features.
KeywordsAdaptive resonance theory Embedding field theory Minimal method anatomy Unsupervised learning Self-defining feature set Set-membership theory
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. The author is indebted to Dr. Richard B. Wells, emeritus professor at the University of Idaho, for valuable suggestions concerning the direction of this research. He would also like to thank Dr. Terence Soule, Dr. Eric Wolbrecht and Dr. Onesmo B. Balemba, all of University of Idaho, for helpful encouragement and criticisms.
- 12.Gardner EP, Martin JH (2000) Coding of sensory information. In: Kandel Eric R, Schwartz James H, Jessell Thomas M (eds) Principles of neural science, vol 4. McGraw-Hill, New York, pp 411–429Google Scholar
- 15.Grossberg S (1982) Classical and instrumental learning by neural networks. In: Cohen RS, Wartofsky MW (eds) Studies in mind and brain: neural principles of learning, perception, development, cognition, and motor control. D. Reidel Publishing Company, Dordrecht, pp 65–156. doi: 10.1007/978-94-009-7758-73 CrossRefGoogle Scholar