A Deeper Look into Aleatoric and Epistemic Uncertainty Disentanglement

Matias Valdenegro-Toro, Daniel Saromo Mori

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

17 Citas (Scopus)

Resumen

Neural networks are ubiquitous in many tasks, but trusting their predictions is an open issue. Uncertainty quantification is required for many applications, and disentangled aleatoric and epistemic uncertainties are best. In this paper, we generalize methods to produce disentangled uncertainties to work with different uncertainty quantification methods, and evaluate their capability to produce disentangled uncertainties. Our results show that: there is an interaction between learning aleatoric and epistemic uncertainty, which is unexpected and violates assumptions on aleatoric uncertainty, some methods like Flipout produce zero epistemic uncertainty, aleatoric uncertainty is unreliable in the out-of-distribution setting, and Ensembles provide overall the best disentangling quality. We also explore the error produced by the number of samples hyper-parameter in the sampling softmax function, recommending N > 100 samples. We expect that our formulation and results help practitioners and researchers choose uncertainty methods and expand the use of disentangled uncertainties, as well as motivate additional research into this topic.

Idioma originalInglés
Título de la publicación alojadaProceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2022
EditorialIEEE Computer Society
Páginas1508-1516
Número de páginas9
ISBN (versión digital)9781665487399
DOI
EstadoPublicada - 2022
Evento2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2022 - New Orleans, Estados Unidos
Duración: 19 jun. 202220 jun. 2022

Serie de la publicación

NombreIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
Volumen2022-June
ISSN (versión impresa)2160-7508
ISSN (versión digital)2160-7516

Conferencia

Conferencia2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2022
País/TerritorioEstados Unidos
CiudadNew Orleans
Período19/06/2220/06/22

Huella

Profundice en los temas de investigación de 'A Deeper Look into Aleatoric and Epistemic Uncertainty Disentanglement'. En conjunto forman una huella única.

Citar esto