Bayesian Linear Regression on Deep Representations
Paper i proceeding, 2019

A simple approach to obtaining uncertainty-aware neural networks for regression is to do Bayesian linear regression (BLR) on the representation from the last hidden layer. Recent work [Riquelme et al., 2018, Azizzadenesheli et al., 2018] indicates that the method is promising, though it has been limited to homoscedastic noise. In this paper, we propose a novel variation that enables the method to flexibly model heteroscedastic noise. The method is benchmarked against two prominent alternative methods on a set of standard datasets, and finally evaluated as an uncertainty-aware model in model-based reinforcement learning. Our experiments indicate that the method is competitive with standard ensembling, and ensembles of BLR outperforms the methods we compared to.

uncertainty estimation

bayesian linear regression

deep learning

Författare

John Moberg

Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik, Signalbehandling

Lennart Svensson

Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik, Signalbehandling

Juliano Pinto

Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik, Signalbehandling

Henk Wymeersch

Chalmers, Elektroteknik, Kommunikations- och antennsystem, Kommunikationssystem

Advances in Neural Information Processing Systems 32

4th workshop on Bayesian Deep Learning (NeurIPS 2019)
Vancouver, Canada,

Styrkeområden

Informations- och kommunikationsteknik

Ämneskategorier

Datavetenskap (datalogi)

Mer information

Senast uppdaterat

2021-04-16