About

Publications

In computational learning theory, authors are listed alphabetically.

Learning Theory

Efficient Robust Learning at the Information-Theoretic Limit

Adam R. Klivans, Konstantinos Stavropoulos, Sergei Tikhonov, Arsen Vasilyan

arXiv preprint, 2026

We give the first efficient algorithm to achieve information-theoretically optimal error under nasty noise, given an ERM oracle or L2 polynomial approximations for the concept class.

Paper

Learning Theory

Proper Agnostic Learning of Functions of Halfspaces

Sergei Tikhonov, Arsen Vasilyan

NeurIPS 2026Spotlight [TOP 3%]

We created the first algorithm for properly agnostically learning Boolean functions of halfspaces under Gaussian marginals.

Paper Poster