Optimal mistake bounds achieved for online learning with randomization

Optimal Randomized Proper Online Learning

Machine Learning

Summary

When computers learn from data step-by-step without knowing the future, they sometimes make mistakes. This paper shows how to minimize those mistakes to the best possible level for a certain type of learning that picks solutions only from a fixed set and uses random choices. The authors improved the known math to show fewer mistakes happen as the number of learning steps grows. This helps understand the limits of such learning methods in the worst cases.

What this means in practice

A theory result. No direct application yet.

Authors

Zachary Chase, Idan Mehalel

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.