Empirical comparison of ten BNN inference methods shows test log-likelihood can mislead on uncertainty quality and that posterior-structure innovations do not necessarily yield high-quality approximations.
Multiplicative Normalizing Flows for Variational Bayesian Neural Networks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
We reinterpret multiplicative noise in neural networks as auxiliary random variables that augment the approximate posterior in a variational setting for Bayesian neural networks. We show that through this interpretation it is both efficient and straightforward to improve the approximation by employing normalizing flows while still allowing for local reparametrizations and a tractable lower bound. In experiments we show that with this new approximation we can significantly improve upon classical mean field for Bayesian neural networks on both predictive accuracy as well as predictive uncertainty.
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Quality of Uncertainty Quantification for Bayesian Neural Network Inference
Empirical comparison of ten BNN inference methods shows test log-likelihood can mislead on uncertainty quality and that posterior-structure innovations do not necessarily yield high-quality approximations.