Robust AI Defense
Sibylle and Kyle discuss the challenge of creating neural networks that can accurately classify data while being resistant to adversarial attacks. They explore the concept of measuring confidence through clustering and the difficulty of estimating the lipsticks constant. The chapter highlights the ongoing research in this area and the potential for computing guarantees against successful adversarial attacks.In this clip
From this podcast
Data Skeptic
Matrix Factorization For k-Means
Related Questions