K. Hirozawa and Y. Osana (Japan)
Chaotic Neural Network, Time Delay, Dynamic Associa tion
In this paper, we propose a Chaotic Neural Network with Time Delay term for Sequential Patterns (CNNTDSP). The proposed model is based on the conventional chaotic neural network and has two types of connection weights; (1) nor mal weights for hetero associations and (2) weights with time delay for auto associations. The proposed model deal with the sequential patterns. In the proposed model, asso ciations of the sequential pattern in short term and dynamic associations between sequential patterns in long term are realized. We carried out a series of computer experiments and confirmed the effectiveness of the proposed model.
Important Links:
Go Back
