On the Hardness of Robust Classification
Pascale Gourdeau, Varun Kanade, Marta Kwiatkowska, James Worrell
–Neural Information Processing Systems
We start with two negative results. We show that no non-trivial concept class can be robustly learned in the distribution-free setting against an adversary who can perturb just a single input bit.
Neural Information Processing Systems
Nov-17-2025, 13:07:22 GMT
- Country:
- Europe > United Kingdom
- England > Oxfordshire > Oxford (0.05)
- North America > Canada (0.04)
- Europe > United Kingdom
- Technology:
- Information Technology
- Artificial Intelligence > Machine Learning (1.00)
- Data Science (1.00)
- Security & Privacy (0.68)
- Information Technology