Evaluation of neural network performance and generalisation using thresholding functions
- 120 Downloads
The application of a simple thresholding technique to help assess the satisfactory performance of classification networks formed from Multi-Layer Perceptron (MLP) artificial neural networks (ANNs) is discussed. Both conventional Maximum Likelihood and Bayesian Evidence based training paradigms were implemented. Firstly a simulated data set drawn from a two-dimensional Gaussian distribution was investigated to illustrate the physical significance of the threshold plots compared to the classifier output probability contours. Secondly a real world application data set comprising of low-frequency vibration measurements on an aircraft wing (a GNAT trainer) is considered. It is demonstrated that simple threshold based plots applied to classifier network outputs may provide a simple yet powerful technique to aid in the rejection of poorly regularised network structures.
KeywordsNeural network training and generalisation Thresholding
This work was supported by EPSRC grant number GR/R96415/01 in association with DSTL Farnborough who are acknowledged for sample provision and assistance with data collection. The authors gratefully acknowledge the use of the software package NETLAB developed by Ian Nabney of Aston University [http://www.ncrg.aston.ac.uk/netlab/].
- 2.Zhang GP (2000) Neural networks for classification: a survey. IEEE Trans Systems, Man and Cybernetics-Part C: Applications and Reviews 30(4):451–462Google Scholar
- 6.Masnata A, Sunseri M (1996) Neural network classification of flaws detected by ultrasonic means. NDT&E Int 29(2):97–93Google Scholar
- 12.Bishop CM (1995) Neural networks for pattern recognition. Oxford University Press, New YorkGoogle Scholar
- 15.MacKay DJC (1992) A practical Bayesian framework for back-propagation networks. Neural Comput 4:448–472Google Scholar
- 16.Worden K, Tomlinson GR (2001) Nonlinearity in structural dynamics. Institute of Physics PublishingGoogle Scholar