Neural Network Ensembles and Variational Inference Revisited

Ensembling methods and variational inference provide two orthogonal methods for obtaining reliable predictive uncertainty estimates for neural networks. In this work we compare and combine these approaches finding that: i) variational inference outperforms ensembles of neural networks, and ii) ensembled versions of variational inference bring further improvements. The first finding appears at odds with previous work (Lakshminarayananet al., 2017), but we show that the previous results were due to an ambiguous experimental protocol in which the model and inference method were simultaneously changed.

View the paper