Abstract
We introduce a new method for quantifying prediction uncertainty in regression neural networks using evidential likelihood-based inference. The method is based on the Gaussian approximation of the likelihood function and the linearization of the network output with respect to the weights. Prediction uncertainty is described by a random fuzzy set inducing a predictive belief function. Preliminary experiments suggest that the approximations are very accurate and that the method allows for conservative uncertainty-aware predictions.
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
References
Abdar, M., et al.: A review of uncertainty quantification in deep learning: techniques, applications and challenges. Inf. Fusion 76, 243–297 (2021)
Couso, I., Sánchez, L.: Upper and lower probabilities induced by a fuzzy random variable. Fuzzy Sets Syst. 165(1), 1–23 (2011)
Denœux, T.: A \(k\)-nearest neighbor classification rule based on Dempster-Shafer theory. IEEE Trans. Syst. Man Cybern. 25(05), 804–813 (1995)
Denœux, T.: A neural network classifier based on Dempster-Shafer theory. IEEE Trans. Syst. Man Cybern. A 30(2), 131–150 (2000)
Denœux, T.: An evidential neural network model for regression based on random fuzzy numbers. In: Le Hégarat-Mascle, S., Bloch, I., Aldea, E. (eds.) BELIEF 2022. LNCS, vol. 13506, pp. 57–66. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-17801-6_6
Denœux, T.: Quantifying prediction uncertainty in regression using random fuzzy sets: the ENNreg model. IEEE Trans. Fuzzy Syst. 31, 3690–3699 (2023)
Denœux, T.: Reasoning with fuzzy and uncertain evidence using epistemic random fuzzy sets: general framework and practical models. Fuzzy Sets Syst. 453, 1–36 (2023)
Denœux, T.: Uncertainty quantification in logistic regression using random fuzzy sets and belief functions. Int. J. Approximate Reasoning 168, 109159 (2024)
Denœux, T., Kanjanatarakul, O., Sriboonchitta, S.: A new evidential k-nearest neighbor rule based on contextual discounting with partially supervised learning. Int. J. Approximate Reasoning 113, 287–302 (2019)
Hüllermeier, E., Waegeman, W.: Aleatoric and epistemic uncertainty in machine learning: an introduction to concepts and methods. Mach. Learn. 110(3), 457–506 (2021)
Kanjanatarakul, O., Sriboonchitta, S., Denœux, T.: Forecasting using belief functions: an application to marketing econometrics. Int. J. Approximate Reasoning 55(5), 1113–1128 (2014)
Kanjanatarakul, O., Sriboonchitta, S., Denœux, T.: Prediction of future observations using belief functions: a likelihood-based approach. Int. J. Approximate Reasoning 72, 71–94 (2016)
Nguyen, H.T.: On random sets and belief functions. J. Math. Anal. Appl. 65, 531–542 (1978)
Sprott, D.A.: Statistical Inference in Science. Springer, Berlin (2000). https://doi.org/10.1007/b98955
Zadeh, L.A.: The concept of a linguistic variable and its application to approximate reasoning -I. Inf. Sci. 8, 199–249 (1975)
Zadeh, L.A.: Fuzzy sets as a basis for a theory of possibility. Fuzzy Sets Syst. 1, 3–28 (1978)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Denœux, T. (2024). Uncertainty Quantification in Regression Neural Networks Using Likelihood-Based Belief Functions. In: Bi, Y., Jousselme, AL., Denoeux, T. (eds) Belief Functions: Theory and Applications. BELIEF 2024. Lecture Notes in Computer Science(), vol 14909. Springer, Cham. https://doi.org/10.1007/978-3-031-67977-3_5
Download citation
DOI: https://doi.org/10.1007/978-3-031-67977-3_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-67976-6
Online ISBN: 978-3-031-67977-3
eBook Packages: Computer ScienceComputer Science (R0)