Introduction to statistical pattern recognition (2nd ed.)
Introduction to statistical pattern recognition (2nd ed.)
Fundamentals of neural networks: architectures, algorithms, and applications
Fundamentals of neural networks: architectures, algorithms, and applications
Pattern classification: a unified view of statistical and neural approaches
Pattern classification: a unified view of statistical and neural approaches
Associative dynamics in a chaotic neural network
Neural Networks
Neural Networks for Pattern Recognition
Neural Networks for Pattern Recognition
Pattern Recognition and Neural Networks
Pattern Recognition and Neural Networks
Periodicity and stability issues of a chaotic pattern recognition neural network
Pattern Analysis & Applications
Desynchronizing a Chaotic Pattern Recognition Neural Network to Model Inaccurate Perception
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
| Hi-index | 0.00 |
Chaotic Pattern Recognition (PR) is a relatively new sub-fieldof PR in which a system, which demonstrates chaotic behavior undernormal conditions, resonates when it is presented with a patternthat it is trained with. The Adachi Neural Network (AdNN) is aclassic neural structure which has been proven to demonstrate thephenomenon of Associative Memory (AM). In their pioneering paper[1,2] , Adachi and his co-authors showed that the AdNN alsoemanates periodic outputs on being exposed to trained patterns.This was later utilized by Calitoiu et al [4,5] to designsystems which possibly possessed PR capabilities. In this paper, weshow that the previously reported properties of the AdNN do notadequately describe the dynamics of the system. Rather, although itpossesses far more powerful PR and AM properties than was earlierknown, it goes through a spectrum of characteristics as one of itscrucial parameters, α , changes. As α increases, the AdNN which is first an AM becomequasi -chaotic. The system is then distinguished by twophases which really do not have clear boundaries of demarcation. Inthe former of these phases it is quasi -chaotic for somepatterns and periodic for others. In the latter of these, itexhibits properties that have been unknown (or rather, unreported)till now, namely, a PR capability (which even recognizes masked oroccluded patterns) in which the network resonates sympatheticallyfor trained patterns while it is quasi -chaotic foruntrained patterns. Finally, the system becomes completelyperiodic. All these results are, to the best of our knowledge,novel.