In this paper, we discuss an approach for designing the computational neural network, which is mainly composed of a hardlimiter neuron, a updated neuron, and a search function neuron, to solve some computational problems. The computation-by-search (CBS) scheme can effectively solve some complicated problems in the condition that their search functions can be easily obtainable by some existing neural networks. The convergence of the suggested neural networks to achieve the solution are discussed and analyzed. Both theoretical analyses and simulated results show that the proposed neural network can effectively solve the complicated computational problems such that they belong to the rational functions or their inverse functions can be easily implemented by using an existing network.
|Number of pages||5|
|Publication status||Published - 1995 Dec 1|
|Event||Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6) - Perth, Aust|
Duration: 1995 Nov 27 → 1995 Dec 1
|Other||Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6)|
|Period||95-11-27 → 95-12-01|
All Science Journal Classification (ASJC) codes