Uncertainty quantification methods for neural networks pattern recognition

Tolo, Silvia and Santhosh, T. V. and Vinod, Gopika and Oparaji, Uchenna and Patelli, Edoardo; (2018) Uncertainty quantification methods for neural networks pattern recognition. In: 2017 IEEE Symposium Series on Computational Intelligence. Institute of Electrical and Electronics Engineers Inc., USA. ISBN 9781538627259

[img]
Preview
Text (Tolo-etal-SSCI2017-Uncertainty-quantification-methods-for-neural-networks-pattern-recognition)
Tolo_etal_SSCI2017_Uncertainty_quantification_methods_for_neural_networks_pattern_recognition.pdf
Accepted Author Manuscript

Download (531kB)| Preview

    Abstract

    On-line monitoring techniques have attracted increasing attention as a promising strategy for improving safety, maintaining availability and reducing the cost of operation and maintenance. In particular, pattern recognition tools such as artificial neural networks are today largely adopted for sensor validation, plant component monitoring, system control, and fault-diagnostics based on the data acquired during operation. However, classic artificial neural networks do not provide an error context for the model response, whose robustness remains thus difficult to estimate. Indeed, experimental data generally exhibit a time/space-varying behaviour and are hence characterized by an intrinsic level of uncertainty that unavoidably affects the performance of the tools adopted and undermines the accuracy of the analysis. For this reason, the propagation of the uncertainty and the quantification of the so called margins of uncertainty in output are crucial in making risk-informed decision. The current study presents a comparison between two different approaches for the quantification of uncertainty in artificial neural networks. The first technique presented is based on the error estimation by a series association scheme, the second approach couples Bayesian model selection technique and model averaging into a unified framework. The efficiency of these two approaches are analysed in terms of their computational cost and predictive performance, through their application to a nuclear power plant fault diagnosis system.