Soft Computing

, Volume 21, Issue 9, pp 2385–2393

On randomization of neural networks as a form of post-learning strategy

Methodologies and Application

DOI: 10.1007/s00500-015-1949-1

Cite this article as:
Kapanova, K.G., Dimov, I. & Sellier, J.M. Soft Comput (2017) 21: 2385. doi:10.1007/s00500-015-1949-1

Abstract

Today artificial neural networks are applied in various fields—engineering, data analysis, robotics. While they represent a successful tool for a variety of relevant applications, mathematically speaking they are still far from being conclusive. In particular, they suffer from being unable to find the best configuration possible during the training process (local minimum problem). In this paper, we focus on this issue and suggest a simple, but effective, post-learning strategy to allow the search for improved set of weights at a relatively small extra computational cost. Therefore, we introduce a novel technique based on analogy with quantum effects occurring in nature as a way to improve (and sometimes overcome) this problem. Several numerical experiments are presented to validate the approach.

Keywords

Neural networks Quantum randomness Training strategy Function approximation 

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  1. 1.IICT, Bulgarian Academy of SciencesSofiaBulgaria

Personalised recommendations