CUED Publications database

High-quality prediction intervals for deep learning: A distribution-free, ensembled approach

Pearce, T and Zaki, M and Brintrup, A and Neely, A (2018) High-quality prediction intervals for deep learning: A distribution-free, ensembled approach. In: UNSPECIFIED pp. 6473-6482..

Full text not available from this repository.

Abstract

©35th International Conference on Machine Learning, ICML 2018.All Rights Reserved. This paper considers the generation of prediction intervals (PIs) by neural networks for quantifying uncertainty in regression tasks. It is axiomatic that high-quality PIs should be as narrow as possible, whilst capturing a specified portion of data. We derive a loss function directly from this axiom that requires no distributional assumption. We show how its form derives from a likelihood principle, that it can be used with gradient descent, and that model uncertainty is accounted for in ensembled form. Benchmark experiments show the method outperforms current state-of-the-art uncertainty quantification methods, reducing average PI width by over 10%.

Item Type: Conference or Workshop Item (UNSPECIFIED)
Subjects: UNSPECIFIED
Divisions: Div E > Strategy and Policy
Div E > Manufacturing Systems
Depositing User: Cron Job
Date Deposited: 11 Dec 2018 01:36
Last Modified: 24 Oct 2019 14:38
DOI: