2 papers
stat.ML2020
Improving predictions of Bayesian neural nets via local linearization
Alexander Immer, Maciej Korzepa, Matthias Bauer
The generalized Gauss-Newton (GGN) approximation is often used to make practical Bayesian deep learning approaches scalable by replacing a second order derivative with a product of…
stat.ML2019
Approximate Inference Turns Deep Networks into Gaussian Processes
Mohammad Emtiyaz Khan, Alexander Immer, Ehsan Abedi +1
Deep neural networks (DNN) and Gaussian processes (GP) are two powerful models with several theoretical connections relating them, but the relationship between their training metho…