Re:Interview with Xavier Leroy
by bagnalla on Friday December 08, 2017 @08:54PM
(#15883772)
Verification of machine learning models, especially w.r.t. robustness to so-called "adversarial examples" [1], is a major research topic at the moment. There are certainly people worrying about it in the machine learning community, but it appears to be a very difficult problem.
[1] https://arxiv.org/abs/1312.6199