paper
referenced-only
2019
paper:engstrom-2019-adversarialAdversarial Robustness as a Prior for Learned Representations
ByLogan Engstrom·Andrew Ilyas·Shibani Santurkar·Dimitris Tsipras·Brandon Tran·A. Ma̧dry
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