Maria-Florina Balcan, Dan DeBlasio, Travis Dick, Carl Kingsford, Tuomas Sandholm, and Ellen Vitercik (2024) Journal version of "How much data is sufficient to learn high-performing algorithms?". J. ACM 71(5):32, pages 1--58 .

The journal version of our 2021 STOC paper "How much data is sufficient to learn high-performing algorithms" has been accepted to the Journal of the ACM.

The STOC 2021 version can be found here.

See also this post.

DOI | Preprint