AAAI 2023 Distinguished Paper - Explainable AI, Differential Privacy, and Certified Robustness

[nhathai phan]

Dear All, 

On behalf of our team, I am happy to share that paper "XRand: Differentially Private Defense against Explanation-Guided Attacks" was selected as one of the twelve AAAI 2023 Distinguished Papers. AAAI Conference Paper Awards and Recognition honors papers that exemplify the highest standards in technical contribution and exposition.

In this paper, we draw the first theoretical connection between Explainable AI, Differential Privacy, and Certified Robustness (in both training and inference). That brings Trustworthy AI closer to real-world adoption.

I am looking forward to seeing you in #AAAI2023. Please stop by our presentation in the APP: Security Session, Sunday, February 12, 3:45-5:00pm - Room 140A.

Best regards,

Hai Phan

YWCC, NJIT