Abstract
Most machine learning methods used for regression explicitly or implicitly assume a symmetric loss function. However, recently an increasing number of problem domains require loss functions that are asymmetric in the sense that the costs for over- or under-predicting the target value may differ. This paper discusses theoretical foundations of handling asymmetric loss functions, and describes and evaluates simple methods which might be used to offset the effects of asymmetric losses. While these methods are applicable to any problem where an asymmetric loss is used, our work derives its motivation from the area of predictive maintenance, which is often characterized by a small number of training samples (in case of failure prediction) or monetary cost-based, mostly non-convex, loss functions.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
The datasets used are auto93.arff, autoMpg.arff, autoPrice.arff, cloud.arff, cpu.arff, echoMonths.arff, elevators.arff, housing.arff, meta.arff, pyrim.arff, strike.arff, triazines.arff, and veteran.arff.
References
Calvo, B., Santafe, G.: scmamp: Statistical comparison of multiple algorithms in multiple problems. R J. 8(1), 248–256 (2016)
Christoffersen, P.F., Diebold, F.X.: Optimal prediction under asymmetric loss. Econom. Theor. 13, 808–817 (1997)
Demšar, J.: Statistical comparisons of classifiers over multiple data sets. J. Mach. Learn. Res. 7, 1–30 (2006)
Elkan, C.: The foundations of cost-sensitive learning. In: Proceedings of the 17th International Joint Conference on Artificial Intelligence (IJCAI 2001), vol. 2. pp. 973–978. Morgan Kaufmann Publishers Inc, San Francisco, CA, USA (2001)
Frank, E., Wang, Y., Inglis, S., Holmes, G., Witten, I.H.: Using model trees for classification. Mach. Learn. 32(1), 63–76 (1998)
Granger, C.W.J.: Prediction with a generalized cost of error function. Oper. Res. Q. 20(2), 199–207 (1969)
Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The Weka data mining software: An update. SIGKDD Explor. 11(1), 10–18 (2009)
Hernández-Orallo, J.: Probabilistic reframing for cost-sensitive regression. ACM Trans. Knowl. Discov. Data 8(4), 17.1–17.55 (2014)
Léger, C., Romano, J.P.: Bootstrap choice of tuning parameters. Annal. Inst. Stat. Math. 42(4), 709–735 (1990)
Lichman, M.: UCI machine learning repository (2013). http://archive.ics.uci.edu/ml
McCullough, B.: Optimal prediction with a general loss function. J. Comb. Inf. Syst. Sci. 25(14), 207–221 (2000)
Zhao, H., Sinha, A.P., Bansal, G.: An extended tuning method for cost-sensitive regression and forecasting. Decis. Support Syst. 51(3), 372–383 (2011)
Acknowledgements
This work was supported by the German Federal Ministry of Education and Research (BMBF) under the “An Optic’s Life” project (no. 16KIS0025). Thanks to the reviewers of this and a previous version of the paper, in particular for the pointers to prior work.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Tolstikov, A., Janssen, F., Fürnkranz, J. (2017). Evaluation of Different Heuristics for Accommodating Asymmetric Loss Functions in Regression. In: Yamamoto, A., Kida, T., Uno, T., Kuboyama, T. (eds) Discovery Science. DS 2017. Lecture Notes in Computer Science(), vol 10558. Springer, Cham. https://doi.org/10.1007/978-3-319-67786-6_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-67786-6_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-67785-9
Online ISBN: 978-3-319-67786-6
eBook Packages: Computer ScienceComputer Science (R0)