Completed
PAC-Bayes bounds on deterministic classifiers
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
PAC-Bayesian Approaches to Understanding Generalization in Deep Learning - Gintare Dziugaite
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Setup
- 3 Outline
- 4 PAC-Bayes yields risk bounds for Gibbs classifiers
- 5 PAC-Bayes generalization bounds
- 6 PAC-Bayes bounds on deterministic classifiers
- 7 Recap: Towards a nonvacuous bound on SGD
- 8 Can we exploit optimal priors?
- 9 Distribution-dependent priors (Lever et al. 2010)
- 10 Empirical evaluation of Lever et al.'s bounds
- 11 Distribution-dependent approximations of optimal priors via privacy
- 12 A question of interpretation
- 13 Data-dependent oracle priors for neural networks
- 14 Coupled data-dependent approximate oracle priors and posteriors
- 15 Gaussian network bounds for Coupled data-dependent priors
- 16 Oracle access to optimal prior covariance
- 17 Directly optimizing Variational data-dependent PAC-Bayes generalization bound.
- 18 Recap and Conclusion
