Abstract
A method is proposed for determination and adaptive alteration of artificial neural network (ANN) structures. Not only the weights but also the structure is altered adaptively. From the engineering viewpoint, such an adaptation will be beneficial, for example, for accommodation to faults which requires different structures of the ANN. The salient point of the proposed method is that it can alter the ANN structure by adjusting a single parameter; we do not need to worry about which particular nodes to be removed or where and how many nodes to be added. The method is based on genetic algorithms (GA) which emulate evolution. A fitness function and a coding system of ANN structures on chromosomes are also proposed which are appropriate for optimization of the ANN structures.
| Original language | English |
|---|---|
| Pages (from-to) | 1690-1695 |
| Number of pages | 6 |
| Journal | Proceedings of the IEEE International Conference on Systems, Man and Cybernetics |
| Volume | 3 |
| Publication status | Published - 1996 |
| Event | Proceedings of the 1996 IEEE International Conference on Systems, Man and Cybernetics. Part 3 (of 4) - Beijing, China Duration: Oct 14 1996 → Oct 17 1996 |
All Science Journal Classification (ASJC) codes
- Control and Systems Engineering
- Hardware and Architecture