Training neural networks using Taguchi methods: overcoming interaction problems.
Maxwell, Grant M.
MetadataShow full item record
VISWANATHAN, A., MACLEOD, C., MAXWELL, G. and KALIDINDI, S. 2005. Training neural networks using Taguchi methods: overcoming interaction problems. Lecture notes in computer science [online], 3697, Proceedings of the 15th international conference on artifical neural networks (ICANN 2005): formal models and their applications, 11-15 September 2005, Warsaw, Poland, part 2, pages 103-108. Available from: https://dx.doi.org/10.1007/11550907_17
Taguchi Methods (and other orthogonal arrays) may be used to train small Artificial Neural Networks very quickly in a variety of tasks. These include, importantly, Control Systems. Previous experimental work has shown that they could be successfully used to train single layer networks with no difficulty. However, interaction between layers precluded the successful reliable training of multi-layered networks. This paper describes a number of successful strategies which may be used to overcome this problem and demonstrates the ability of such networks to learn non-linear mappings.