Laplacian Segmentation Networks Improve Epistemic Uncertainty Quantification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zepf, Kilian, Wanna, Selma, Miani, Marco, Moore, Juston, Frellsen, Jes, Hauberg, Søren, Warburg, Frederik, Feragen, Aasa
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916332047433728
author Zepf, Kilian
Wanna, Selma
Miani, Marco
Moore, Juston
Frellsen, Jes
Hauberg, Søren
Warburg, Frederik
Feragen, Aasa
author_facet Zepf, Kilian
Wanna, Selma
Miani, Marco
Moore, Juston
Frellsen, Jes
Hauberg, Søren
Warburg, Frederik
Feragen, Aasa
contents Image segmentation relies heavily on neural networks which are known to be overconfident, especially when making predictions on out-of-distribution (OOD) images. This is a common scenario in the medical domain due to variations in equipment, acquisition sites, or image corruptions. This work addresses the challenge of OOD detection by proposing Laplacian Segmentation Networks (LSN): methods which jointly model epistemic (model) and aleatoric (data) uncertainty for OOD detection. In doing so, we propose the first Laplace approximation of the weight posterior that scales to large neural networks with skip connections that have high-dimensional outputs. We demonstrate on three datasets that the LSN-modeled parameter distributions, in combination with suitable uncertainty measures, gives superior OOD detection.
format Preprint
id arxiv_https___arxiv_org_abs_2303_13123
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Laplacian Segmentation Networks Improve Epistemic Uncertainty Quantification
Zepf, Kilian
Wanna, Selma
Miani, Marco
Moore, Juston
Frellsen, Jes
Hauberg, Søren
Warburg, Frederik
Feragen, Aasa
Computer Vision and Pattern Recognition
Machine Learning
Image segmentation relies heavily on neural networks which are known to be overconfident, especially when making predictions on out-of-distribution (OOD) images. This is a common scenario in the medical domain due to variations in equipment, acquisition sites, or image corruptions. This work addresses the challenge of OOD detection by proposing Laplacian Segmentation Networks (LSN): methods which jointly model epistemic (model) and aleatoric (data) uncertainty for OOD detection. In doing so, we propose the first Laplace approximation of the weight posterior that scales to large neural networks with skip connections that have high-dimensional outputs. We demonstrate on three datasets that the LSN-modeled parameter distributions, in combination with suitable uncertainty measures, gives superior OOD detection.
title Laplacian Segmentation Networks Improve Epistemic Uncertainty Quantification
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2303.13123