paper
referenced-only
2018
paper:gilmer-2018-adversarialAdversarial Spheres
ByJ. Gilmer·Luke Metz·Fartash Faghri·S. Schoenholz·M. Raghu·M. Wattenberg+1 more
Similar preprints — Semantic Scholar
Cited by (1)
- Adversarial Examples Are Not Bugs, They Are Superposition
Superposition causally contributes to adversarial vulnerability, and the link runs in both directions: in a toy ReLU autoencoder (n=100 features, m=20 hidden units), raising feature sparsity raises fe