Completed
Randomization as a defense?
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Provable Robustness Beyond Bound Propagation
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Adversarial attacks on deep learning
- 3 Why should we care?
- 4 Adversarial robustness
- 5 How to we strictly upper bound the maximization?
- 6 This talk
- 7 What causes adversarial examples?
- 8 Randomization as a defense?
- 9 Visual intuition of randomized smoothing
- 10 The randomized smoothing guarantee
- 11 Proof of certified robustness (cont)
- 12 Caveats (a.k.a. the fine print)
- 13 Comparison to previous SOTA on CIFAR10
- 14 Performance on ImageNet