# Practical Deep Learning with Bayesian Principles

@article{Osawa2019PracticalDL, title={Practical Deep Learning with Bayesian Principles}, author={Kazuki Osawa and Siddharth Swaroop and Anirudh Jain and Runa Eschenhagen and Richard E. Turner and Rio Yokota and Mohammad Emtiyaz Khan}, journal={ArXiv}, year={2019}, volume={abs/1906.02506} }

Bayesian methods promise to fix many shortcomings of deep learning, but they are impractical and rarely match the performance of standard methods, let alone improve them. In this paper, we demonstrate practical training of deep networks with natural-gradient variational inference. By applying techniques such as batch normalisation, data augmentation, and distributed training, we achieve similar performance in about the same number of epochs as the Adam optimiser, even on large datasets such as… CONTINUE READING

13

Twitter Mentions

#### Supplemental Code

GITHUB REPO

Via Papers with Code

Contains code for the NeurIPS 2019 paper "Practical Deep Learning with Bayesian Principles"

#### Citations

##### Publications citing this paper.

SHOWING 1-10 OF 32 CITATIONS

## Efficient and Scalable Bayesian Neural Nets with Rank-1 Factors

VIEW 1 EXCERPT

CITES BACKGROUND

## Convergence Rates of Variational Inference in Sparse Deep Learning

VIEW 1 EXCERPT

CITES METHODS

## Scalable and Practical Natural Gradient for Large-Scale Deep Learning

VIEW 2 EXCERPTS

CITES METHODS

## Approximate Inference Turns Deep Networks into Gaussian Processes

VIEW 3 EXCERPTS

CITES METHODS

## On the Expressiveness of Approximate Inference in Bayesian Neural Networks

VIEW 1 EXCERPT

CITES METHODS

## Conservative Uncertainty Estimation By Fitting Prior Networks

VIEW 8 EXCERPTS

CITES METHODS & BACKGROUND

## Continual Deep Learning by Functional Regularisation of Memorable Past

VIEW 2 EXCERPTS

CITES BACKGROUND

### FILTER CITATIONS BY YEAR

### CITATION STATISTICS

**2**Highly Influenced Citations**Averaged 16 Citations**per year from 2019 through 2020

#### References

##### Publications referenced by this paper.

SHOWING 1-10 OF 59 REFERENCES

## Noisy Natural Gradient as Variational Inference

VIEW 3 EXCERPTS

## A Scalable Laplace Approximation for Neural Networks

VIEW 1 EXCERPT

## Ensemble learning in Bayesian neural networks

VIEW 1 EXCERPT