Scale learning rate by 1/sqrt(f) when sampling rows, Microsoft finds

JohnCLangford · x · 2026-08-14

John Langford shares a rule of thumb: using a fraction f of rows implies scaling the learning rate by 1/sqrt(f); joint tuning yields statistical convergence gains of roughly 1 centinat on some models, part of the Dion optimizer's statistical improvements.

Related event: Microsoft researchers improve convergence with row norm and sampling tuning(2 posts)→

Original post →

More from Research

Research channel →