Laplace Redux - Effortless Bayesian Deep Learning

By Erik A. Daxberger, Agustinus Kristiadi, Alexander Immer, Runa Eschenhagen, Matthias Bauer, Philipp Hennig. Neural Information Processing Systems, 2021

467 citations recorded by the source corpus.

Bayesian formulations of deep learning have been shown to have compelling theoretical properties and offer practical functional benefits, such as improved predictive uncertainty quantification and model selection. The Laplace approximation (LA) is a classic, and arguably the simplest family of approximations for the intractable posteriors of deep neural networks. Yet, despite its simplicity, the LA is not as popular as alternatives like variational Bayes or deep ensembles. This may be due to assumptions that the LA is expensive due to the involved Hessian computation, that it is difficult to implement, or that it yields inferior results. In this work we show that these are misconceptions: we (i) review the range of variants of the LA including versions with minimal cost overhead; (ii) introduce"laplace", an easy-to-use software library for PyTorch offering user-friendly access to all major flavors of the LA; and (iii) demonstrate through extensive experiments that the LA is competitive with more popular alternatives in terms of performance, while excelling in terms of computational cost. We hope that this work will serve as a catalyst to a wider adoption of the LA in practical deep learning, including in domains where Bayesian approaches are not typically considered at the moment.

View Paper: https://www.semanticscholar.org/paper/8415db10a4e2354b7d57b75848f1485ce5d0e7ce

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

  1. Michael Matena
  2. Colin Raffel
  3. Yi Tay
  4. Mohit Bansal
  5. Timo Schick
  6. Gabriel Ilharco
  7. Mostafa Dehghani
  8. Mitchell Wortsman

Citations in the corpus