Optimal Multiclass Online Regret with Arbitrary Label Spaces
arXiv:2610.32339v1 Announce Type: cross Abstract: We determine the optimal regret in agnostic multiclass online learning with arbitrary, possibly infinite, label spaces. For every hypothesis class of multiclass Littlestone dimension $d<\infty$, we prove a regret upper bound of $O(\min\{T,\sqrt{dT}\…
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-10-26 04:00 · arXiv stat.ML
Optimal Multiclass Online Regret with Arbitrary Label Spaces