Konstantin Mishchenko
@konstmish.bsky.social
Research Scientist at Meta Paris
Code generation, math, optimization
Code generation, math, optimization
I think it's much more important we get a better scoring system for matching reviewers and papers. High affinity scores on OpenReview are often misleading. A lot of reviewers complained to me they get random papers from TMLR, and they don't enjoy reviewing as a consequence.
December 13, 2024 at 4:15 PM
I think it's much more important we get a better scoring system for matching reviewers and papers. High affinity scores on OpenReview are often misleading. A lot of reviewers complained to me they get random papers from TMLR, and they don't enjoy reviewing as a consequence.
It's a bit hard to say because this kind of results are still quite new, but one of the most recent papers on the topic, arxiv.org/abs/2410.16249, mentions a conjecture on the optimality of its 1/n^{log₂(1+√ 2)} (not for the last iterate though).
November 27, 2024 at 11:02 PM
It's a bit hard to say because this kind of results are still quite new, but one of the most recent papers on the topic, arxiv.org/abs/2410.16249, mentions a conjecture on the optimality of its 1/n^{log₂(1+√ 2)} (not for the last iterate though).