Sample-based Uncertainty Quantification with a Single Deterministic Neural Network

Takuya Kanazawa, Chetan Gupta

2022

Abstract

Development of an accurate, flexible, and numerically efficient uncertainty quantification (UQ) method is one of fundamental challenges in machine learning. Previously, a UQ method called DISCO Nets has been proposed (Bouchacourt et al., 2016) that trains a neural network by minimizing the so-called energy score on training data. This method has shown superior performance on a hand pose estimation task in computer vision, but it remained unclear whether this method works as nicely for regression on tabular data, and how it competes with more recent advanced UQ methods such as NGBoost. In this paper, we propose an improved neural architecture of DISCO Nets that admits a more stable and smooth training. We benchmark this approach on miscellaneous real-world tabular datasets and confirm that it is competitive with or even superior to standard UQ baselines. We also provide a new elementary proof for the validity of using the energy score to learn predictive distributions. Further, we point out that DISCO Nets in its original form ignore epistemic uncertainty and only capture aleatoric uncertainty. We propose a simple fix to this problem.

Paper Citation

in Harvard Style

Kanazawa T. and Gupta C. (2022). Sample-based Uncertainty Quantification with a Single Deterministic Neural Network. In Proceedings of the 14th International Joint Conference on Computational Intelligence (IJCCI 2022) - Volume 1: NCTA; ISBN 978-989-758-611-8, SciTePress, pages 292-304. DOI: 10.5220/0011546800003332

in Bibtex Style

@conference{ncta22,
author={Takuya Kanazawa and Chetan Gupta},
title={Sample-based Uncertainty Quantification with a Single Deterministic Neural Network},
booktitle={Proceedings of the 14th International Joint Conference on Computational Intelligence (IJCCI 2022) - Volume 1: NCTA},
year={2022},
pages={292-304},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011546800003332},
isbn={978-989-758-611-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Joint Conference on Computational Intelligence (IJCCI 2022) - Volume 1: NCTA
TI - Sample-based Uncertainty Quantification with a Single Deterministic Neural Network
SN - 978-989-758-611-8
AU - Kanazawa T.
AU - Gupta C.
PY - 2022
SP - 292
EP - 304
DO - 10.5220/0011546800003332
PB - SciTePress