Robust Training of Radial Basis Function Neural Networks

0506360 - ÚI 2020 RIV CH eng C - Konferenční příspěvek (zahraniční konf.)
Kalina, Jan - Vidnerová, Petra
Robust Training of Radial Basis Function Neural Networks.
Artificial Intelligence and Soft Computing. Proceedings, Part I. Cham: Springer, 2019 - (Rutkowski, L.; Scherer, R.; Korytkowski, M.; Pedrycz, W.; Tadeusiewicz, R.; Zurada, J.), s. 113-124. Lecture Notes in Computer Science, 11508. ISBN 978-3-030-20911-7. ISSN 0302-9743.
[ICAISC 2019: International Conference on Artificial Intelligence and Soft Computing /18./. Zakopane (PL), 16.06.2019-20.06.2019]
Grant CEP: GA ČR(CZ) GA19-05704S; GA ČR(CZ) GA18-23827S
Institucionální podpora: RVO:67985807
Klíčová slova: Machine learning * Outliers * Robustness * Subset selection * Anomaly detection
Obor OECD: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Radial basis function (RBF) neural networks represent established machine learning tool with various interesting applications to nonlinear regression modeling. However, their performance may be substantially influenced by outlying measurements (outliers). Promising modifications of RBF network training have been available for the classification of data contaminated by outliers, but there remains a gap of robust training of RBF networks in the regression context. A novel robust approach based on backward subsample selection (i.e. instance selection) is proposed and presented in this paper, which searches sequentially for the most reliable subset of observations and finally performs outlier deletion. The novel approach is investigated in numerical experiments and is also applied to robustify a multilayer perceptron. The results on data containing outliers reveal the improved performance compared to conventional approaches.
Trvalý link: http://hdl.handle.net/11104/0297617