Year: 2026
Type: preprint
Abstract: We prove that the optimal expected mistake bound of online learning a function class $\mathcal{H}$ by a randomized proper learning algorithm is $O(\mathtt{L}(\mathcal{H}) \log T)$, where $\mathtt{L}(\mathcal{H})$ is the Littlestone dimension of $\mathcal{H}$ and $T$ is the time horizon. Our result improves upon the previously best known bound of $O(\mathtt{L}(\mathcal{H}) \log^6 T)$ given by Daskalakis and Golowich (STOC 2022), and is optimal up to a universal constant for worst-case classes.
Source: arXiv (Cornell University)
Authors: Chase, Zachary, Mehalel, Idan
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
Funder: European Commission
OpenAlex ID: W7213956648
Cites:
Cited by:
Related to: