Thomas Tanay personal blog
A New Angle on L2 Regularization
Deep neural networks have been shown to be vulnerable to the adversarial example phenomenon: all models tested so far can have their classifications dramatically altered by small image perturbations . The following predictions were for instance made by a state-of-the-art network trained to recognize celebrities :
- Points
- 137
- HN comments
- 10
- Submissions
- 1
- Cited by
- 0 commenters
AI & Machine LearningMath & Algorithms2018
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Jul 2, 2018 | A New Angle on L2 Regularization by lainon | 137 | 10 |
