The Safe Bayesian - Learning the Learning Rate via the Mixability Gap

  title={The Safe Bayesian - Learning the Learning Rate via the Mixability Gap},
  author={Peter Gr{\"u}nwald},
Standard Bayesian inference can behave suboptimally if the model is wrong. We present a modification of Bayesian inference which continues to achieve good rates with wrong models. Our method adapts the Bayesian learning rate to the data, picking the rate minimizing the cumulative loss of sequential prediction by posterior randomization. Our results can also be used to adapt the learning rate in a PAC-Bayesian context. The results are based on an extension of an inequality due to T. Zhang and… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.
11 Citations
27 References
Similar Papers


Publications citing this paper.
Showing 1-10 of 11 extracted citations


Publications referenced by this paper.
Showing 1-10 of 27 references

PAC-Bayesian Supervised Classification

  • O. Catoni
  • Lecture Notes-Monograph Series. IMS,
  • 2007
Highly Influential
5 Excerpts

On information and sufficiency

  • S. Kullback, R. A. Leibler
  • Annals of Mathematical Statistics,
  • 1951
Highly Influential
4 Excerpts

Similar Papers

Loading similar papers…