AI Chat AI Image Generator AI Video Text to Speech

A Note on the Efficient Evaluation of PAC-Bayes Bounds

09/12/2022

∙

by Felix Biggs, et al.

∙

∙

When utilising PAC-Bayes theory for risk certification, it is usually necessary to estimate and bound the Gibbs risk of the PAC-Bayes posterior. Many works in the literature employ a method for this which requires a large number of passes of the dataset, incurring high computational cost. This manuscript presents a very general alternative which makes computational savings on the order of the dataset size.

page 1

page 2

page 3

page 4

research

∙ 08/20/2019

Fast-rate PAC-Bayes Generalization Bounds via Shifted Rademacher Processes

The developments of Rademacher complexity and PAC-Bayesian theory have b...

4 Jun Yang, et al. ∙

research

∙ 10/23/2019

Unifying Variational Inference and PAC-Bayes for Supervised Learning that Scales

Neural Network based controllers hold enormous potential to learn comple...

0 Sanjay Thakur, et al. ∙

research

∙ 02/28/2012

PAC-Bayesian Generalization Bound on Confusion Matrix for Multi-Class Classification

In this work, we propose a PAC-Bayes bound for the generalization risk o...

0 Emilie Morvant, et al. ∙

research

∙ 10/19/2020

PAC^m-Bayes: Narrowing the Empirical Risk Gap in the Misspecified Bayesian Regime

While the decision-theoretic optimality of the Bayesian formalism under ...

0 Warren R. Morningstar, et al. ∙

research

∙ 05/20/2019

PAC-Bayes under potentially heavy tails

We derive PAC-Bayesian learning guarantees for heavy-tailed losses, and ...

0 Matthew J. Holland, et al. ∙

research

∙ 12/18/2021

Data-Driven Reachability analysis and Support set Estimation with Christoffel Functions

We present algorithms for estimating the forward reachable set of a dyna...

0 Alex Devonport, et al. ∙

research

∙ 12/26/2017

Entropy-SGD optimizes the prior of a PAC-Bayes bound: Data-dependent PAC-Bayes priors via differential privacy

We show that Entropy-SGD (Chaudhari et al., 2016), when viewed as a lear...

0 Gintare Karolina Dziugaite, et al. ∙