Universal technique for optimization of neural network training parameters: gasoline near infrared data example
- 110 Downloads
The universal technique of finding optimum training parameters for multi-layer perceptron—such as percentage of samples in a cross-validation set and quantities of training iterations with various initial values—is offered. This technique is aimed at the searching of optimum values of two complex factors depending on accuracy and convergence of a network, and also on the time of its training. Their conventional names are “cross-validation coefficient” and “training iteration coefficient”. Near infrared spectroscopy data for gasoline samples are used to evaluate the efficiency of the method.
KeywordsArtificial neural network (ANN) Multi-layer perceptron (MLP) Parameters optimization Cross-validation coefficient (CVC) Training iteration coefficient (TIC) Gasoline Near infrared (NIR) spectroscopy Density
Balabin Roman is grateful to the ITERA International Group of companies for a nominal scholarship. Authors are grateful to Lumex Ltd. R&D and Production Company (and personally to Demygin SS and Chulyukov OG) for supplying the InfraLUM FT-10 FT-NIR Analyzer.
- 6.Rumelhart DE, Hinton GE, Williams RJ (1986) Learning internal representations by error propagation. In: Rumelhart D, McClelland J (eds) Parallel data processing, vol I, Chap. 8. The MIT Press, Cambridge, pp 318–362Google Scholar
- 10.Balabin RM, Safieva RZ (2008) Motor oil classification by base stock and viscosity based on near infrared (NIR) spectroscopy data. Fuel 87:2745. doi: 10.1016/j.fuel.2008.02.014
- 12.Russian Standard: GOST R 51069-97Google Scholar
- 13.Moody J, Utans J (1992) Principled architecture selection for neural networks: application to corporate bond rating prediction. In: Moody J, Hanson SJ, Lippmann RP (eds) Advances in neural information processing systems, vol IV. Morgan Kaufmann, San Mateo, pp 683–690Google Scholar
- 14.Larsen J, Goutte C (1999) On optimal data split for generalisation estimation and model selection neural networks for signal processing. In: IX Proceedings of the 1999 IEEE signal processing society workshop, pp 225–234Google Scholar
- 22.Allen MP, Tildesley DJ (1989) Computer simulation of liquids. Oxford University Press, New YorkGoogle Scholar