Combining Model and Parameter Uncertainty in Bayesian Neural Networks

03/18/2019
by   Aliaksandr Hubin, et al.
0

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. There are several advantages of using Bayesian approach: Parameter and prediction uncertainty become easily available, facilitating rigid 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 this paper 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. Finally, we show that incorporating model uncertainty via Bayesian model averaging and Bayesian model selection allows to drastically sparsify the structure of BNNs without significant loss of predictive power.

READ FULL TEXT
research
05/01/2023

Variational Inference for Bayesian Neural Networks under Model and Parameter Uncertainty

Bayesian neural networks (BNNs) have recently regained a significant amo...
research
07/17/2019

Subspace Inference for Bayesian Deep Learning

Bayesian inference was once a gold standard for learning with neural net...
research
07/04/2022

Look beyond labels: Incorporating functional summary information in Bayesian neural networks

Bayesian deep learning offers a principled approach to train neural netw...
research
10/21/2020

Incorporating Interpretable Output Constraints in Bayesian Neural Networks

Domains where supervised models are deployed often come with task-specif...
research
05/26/2019

Bayesian Learning of Sum-Product Networks

Sum-product networks (SPNs) are flexible density estimators and have rec...
research
06/03/2019

Bayesian Prior Networks with PAC Training

We propose to train Bayesian Neural Networks (BNNs) by empirical Bayes a...
research
01/29/2020

Bayesian Reasoning with Deep-Learned Knowledge

We access the internalized understanding of trained, deep neural network...

Please sign up or login with your details

Forgot password? Click here to reset