An adversarial example library for constructing attacks, building defenses, and benchmarking both
Role in this project:
ML Engineer Contributions:10 commits, 6 PRs, 8 comments in 1 month
Contributions summary:Jianbo implemented and refined the Boundary Attack++ (BAPP) adversarial example generation method within the CleverHans library. Their contributions involved adding the core BAPP functionality, addressing corner cases, and fixing issues. The user's work focused on modifying and extending the existing attack code and integrating it with the CleverHans framework.
benchmarkingmachine-learningsecurity
Contributions:19 commits, 17 pushes, 1 branch in 1 year 10 months