Journal article

Marginally Calibrated Deep Distributional Regression

Nadja Klein, David J Nott, Michael Stanley Smith

Journal of Computational and Graphical Statistics | American Statistical Association | Published : 2021

Abstract

Deep neural network (DNN) regression models are widely used in applications requiring state-of-the-art predictive accuracy. However, until recently there has been little work on accurate uncertainty quantification for predictions from such models. We add to this literature by outlining an approach to constructing predictive distributions that are “marginally calibrated.” This is where the long run average of the predictive distributions of the response variable matches the observed empirical margin. Our approach considers a DNN regression with a conditionally Gaussian prior for the final layer weights, from which an implicit copula process on the feature space is extracted. This copula proce..

View full abstract

University of Melbourne Researchers

Grants

Awarded by Emmy Noether grant of German Research Foundation (DFG)


Awarded by Singapore Ministry of Education Academic Research Fund Tier 1 grant


Funding Acknowledgements

Nadja Klein acknowledges support through the Emmy Noether grant KL 3037/1-1 of the German Research Foundation (DFG). David Nott was supported by a Singapore Ministry of Education Academic Research Fund Tier 1 grant (R-155-000-189-114).