Bias-regularised neural-network metamodelling of insurance portfolio risk Luo, Wei, Mashrur, Akib, Robles-Kelly, Antonio and Li, Gang 2020, Bias-regularised neural-network metamodelling of insurance portfolio risk, in IJCNN : Proceedings of the 2020 International Joint Conference on Neural Networks, Institute of Electrical and Electronics Engineers (IEEE), Piscataway, N.J., doi: 10.1109/ijcnn48605.2020.9207375. Attached Files Name Description MIMEType Size Downloads Title Bias-regularised neural-network metamodelling of insurance portfolio risk Author(s) Luo, Wei orcid.org/0000-0002-4711-7543 Mashrur, Akib Robles-Kelly, Antonio orcid.org/0000-0002-2465-5971 Li, Gang orcid.org/0000-0003-1583-641X Conference name IJCNN Neural Networks. International Joint Conference (2020 : Glasgow, United Kingdom) Conference location Online : Glasgow, United Kingdom Conference dates 19 - 24 July. 2020 Title of proceedings IJCNN : Proceedings of the 2020 International Joint Conference on Neural Networks Publication date 2020 Total pages 8 Publisher Institute of Electrical and Electronics Engineers (IEEE) Place of publication Piscataway, N.J. Keyword(s) variable annuity metamodelling expected bias percentage error CORE2020 A Summary Deep learning models have attracted considerable attention in metamodelling of financial risks for large insurance portfolios. Those models, however, are generally trained in disregard of the collective nature of the data in the portfolio under study. Consequently, the training procedure often suffers from slow convergence, and the trained model often has poor accuracy. This is particularly evident in the presence of extreme individual contracts. In this paper, we advocate the view that the training of a meta-model for a portfolio should be guided by portfolio-level metrics. In particular, we propose an intuitive loss regulariser that explicitly accounts for the portfolio-level bias. Further, this training regulariser can be easily implemented with the minibatch stochastic gradient descent commonly used in training deep neural networks. Empirical evaluations on both simulated data and a benchmark dataset show that the regulariser yields more stable training, resulting in faster convergence and more reliable portfolio-level risk estimates. ISBN 978-1-7281-6926-2 Language eng DOI 10.1109/ijcnn48605.2020.9207375 Indigenous content off HERDC Research category E1 Full written paper - refereed Copyright notice ©2020, IEEE Persistent URL http://hdl.handle.net/10536/DRO/DU:30143921 Document type: Conference Paper Collections: Faculty of Science, Engineering and Built Environment School of Information Technology Related Links Link Description Connect to published version Go to link with your DU access privileges     Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved. Versions Version Filter Type Mon, 12 Oct 2020, 07:33:35 EST Mon, 12 Oct 2020, 12:52:31 EST Tue, 13 Oct 2020, 04:00:54 EST Tue, 13 Oct 2020, 05:04:38 EST Tue, 13 Oct 2020, 15:15:29 EST Tue, 13 Oct 2020, 15:25:24 EST Filtered Full Citation counts:   Cited 0 times in TR Web of Science Cited 0 times in Scopus Search Google Scholar Access Statistics: 13 Abstract Views, 1 File Downloads  -  Detailed Statistics Created: Mon, 12 Oct 2020, 12:52:31 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.