Multivariate Bayesian Last Layer for Regression: Uncertainty Quantification and Disentanglement

no code implementations • 2 May 2024 • Han Wang, Eiji Kawasaki, Guillaume Damblin, Geoffrey Daniel

We present new Bayesian Last Layer models in the setting of multivariate regression under heteroscedastic noise, and propose an optimization algorithm for parameter learning.

Disentanglement regression +1

Deep Learning reconstruction with uncertainty estimation for $γ$ photon interaction in fast scintillator detectors

no code implementations • 10 Oct 2023 • Geoffrey Daniel, Mohamed Bahi Yahiaoui, Claude Comtat, Sebastien Jan, Olga Kochebina, Jean-Marc Martinez, Viktoriya Sergeyeva, Viatcheslav Sharyy, Chi-Hsun Sung, Dominique Yvon

This article presents a physics-informed deep learning method for the quantitative estimation of the spatial coordinates of gamma interactions within a monolithic scintillator, with a focus on Positron Emission Tomography (PET) imaging.

Packed-Ensembles for Efficient Uncertainty Estimation

1 code implementation • 17 Oct 2022 • Olivier Laurent, Adrien Lafage, Enzo Tartaglione, Geoffrey Daniel, Jean-Marc Martinez, Andrei Bursuc, Gianni Franchi

Deep Ensembles (DE) are a prominent approach for achieving excellent performance on key metrics such as accuracy, calibration, uncertainty estimation, and out-of-distribution detection.

Classifier calibration Image Classification +2