Incorporating Unlabelled Data into Bayesian Neural Networks

04/04/2023
by   Mrinank Sharma, et al.
0

We develop a contrastive framework for learning better prior distributions for Bayesian Neural Networks (BNNs) using unlabelled data. With this framework, we propose a practical BNN algorithm that offers the label-efficiency of self-supervised learning and the principled uncertainty estimates of Bayesian methods. Finally, we demonstrate the advantages of our approach for data-efficient learning in semi-supervised and low-budget active learning problems.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset