Abstract
The behavior of the weight evolution algorithm with deterministic perturbation is presented. The new algorithm evolves the innermost layer of weights in order to escape from local minima. Simulation results have shown that the use of weight evolution with deterministic perturbation leads to an increase in the global convergence percentage, as compared with other traditional methods.
Original language | English |
---|---|
Pages (from-to) | 1039-1041 |
Number of pages | 3 |
Journal | Electronics Letters |
Volume | 36 |
Issue number | 12 |
DOIs | |
Publication status | Published - 8 Jun 2000 |
Externally published | Yes |
ASJC Scopus subject areas
- Electrical and Electronic Engineering