AI RESEARCH
A Framework for Variational Inference of Lightweight Bayesian Neural Networks with Heteroscedastic Uncertainties
arXiv CS.LG
•
ArXi:2402.14532v2 Announce Type: replace Obtaining heteroscedastic predictive uncertainties from a Bayesian Neural Network (BNN) is vital to many applications. Often, heteroscedastic aleatoric uncertainties are learned as outputs of the BNN in addition to the predictive means, however doing so may necessitate adding learnable parameters to the network. In this work, we nstrate that both the heteroscedastic aleatoric and epistemic variance can be embedded into the variances of learned BNN parameters, improving predictive performance for lightweight networks.