Cristin-resultat-ID: 1971502
Sist endret: 22. desember 2021, 13:57
Resultat
Vitenskapelig foredrag
2021

Variational Bayes for inference on model and parameter uncertainty in Bayesian neural networks

Bidragsytere:
  • Aliaksandr Hubin og
  • Geir Olve Storvik

Presentasjon

Navn på arrangementet: The 28th Nordic Conference in Mathematical Statistics
Sted: Tromsø
Dato fra: 21. juni 2021
Dato til: 24. juni 2021

Arrangør:

Arrangørnavn: UiT The Arctic University of Norway

Om resultatet

Vitenskapelig foredrag
Publiseringsår: 2021

Beskrivelse Beskrivelse

Tittel

Variational Bayes for inference on model and parameter uncertainty in Bayesian neural networks

Sammendrag

Bayesian neural networks (BNNs) have recently regained a significant amount of attention in the deep learning community due to the development of scalable approximate Bayesian inference techniques [1]. There are several advantages of using a Bayesian approach: Parameter and prediction uncertainties become easily available, facilitating rigorous statistical analysis. Furthermore, prior knowledge can be incorporated. However so far there have been no scalable techniques capable of combining both model (structural) and parameter uncertainty. In the presented piece of research [2] we introduce the concept of model uncertainty in BNNs and hence make inference in the joint space of models and parameters. Moreover, we suggest an adaptation of a scalable variational inference approach with reparametrization of marginal inclusion probabilities to incorporate the model space constraints. Experimental results on a range of benchmark data sets show that we obtain comparable accuracy results with the competing models, but based on methods that are much more sparse than ordinary BNNs. This is particularly the case in model selection settings, but also within a Bayesian model averaging setting a considerable sparsification is achieved. As expected, model uncertainties give higher, but more reliable uncertainty measures.

Bidragsytere

Aliaksandr Hubin

  • Tilknyttet:
    Forfatter
    ved Statistikk og Data Science ved Universitetet i Oslo
Aktiv cristin-person

Geir Olve Storvik

  • Tilknyttet:
    Forfatter
    ved Statistikk og Data Science ved Universitetet i Oslo
1 - 2 av 2