Project Awesome project awesome

2024 > Under manipulations, are some AI models harder to audit?

Relates the difficulty of black-box audits to the capacity of the targeted models, using the Rademacher complexity.

Package GitHub
Back to Audit Algorithms