Completed
PAC-Bayes bounds on deterministic classifiers
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Studying Generalization in Deep Learning via PAC-Bayes
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 What might generalization theory offer deep learning?
- 3 Barriers to explaining generalization
- 4 PAC-Bayes yields risk bounds for Gibbs classifiers
- 5 PAC-Bayes generalization bounds
- 6 PAC-Bayes bounds on deterministic classifiers
- 7 Distribution-dependent approximations of optimal priors via privacy
- 8 A question of interpretation
- 9 Use SGD to predict SGD
- 10 Data and distribution priors for neural networks
- 11 MNIST Results - Coupled data dependent priors and posteriors
- 12 Oracle access to optimal prior covariance
- 13 Bounds with oracle covariance + ghost sample
- 14 Bounds on 32k samples v 64k samples
- 15 Recap and Conclusion
