GANNET: Design of a neural net for face recognition by Genetic Algorithm

Citation

Hancock PJB & Smith L (1991) GANNET: Design of a neural net for face recognition by Genetic Algorithm. In: Schwefel H & Manner R (eds.) Parallel Problem Solving from Nature: 1st Workshop, PPSN I Dortmund, FRG, October 1–3, 1990 Proceedings. Lecture Notes in Computer Science, 496. 1st Workshop, PPSN I Dortmund, FRG, Dortmund, Germany, 01.10.1990-03.10.1990. Berlin Heidelberg: Springer, pp. 292-296. http://link.springer.com/chapter/10.1007/BFb0029766#; https://doi.org/10.1007/BFb0029766

Abstract
A system that uses a Genetic Algorithm to specify the structure of a back-propagation type neural net is described. The GA can achieve significant improvements in performance over both random and fully connected nets. The efficacy of the evaluation procedure is demonstrated by showing that the system is able to eliminate a small penalty flag despite the high noise levels.