Online Classification with Predictions

TL;DR

This paper introduces an online classification algorithm that leverages predictions of future data to improve learning efficiency, adapting to prediction quality and outperforming traditional worst-case bounds.

Contribution

It presents a novel online learner that integrates predictions, achieving regret bounds that adapt to prediction accuracy and connecting online and transductive learning.

Findings

Expected regret never exceeds worst-case regret

Performance improves with prediction accuracy

Online learning becomes easier with predictable data

Abstract

We study online classification when the learner has access to predictions about future examples. We design an online learner whose expected regret is never worse than the worst-case regret, gracefully improves with the quality of the predictions, and can be significantly better than the worst-case regret when the predictions of future examples are accurate. As a corollary, we show that if the learner is always guaranteed to observe data where future examples are easily predictable, then online learning can be as easy as transductive online learning. Our results complement recent work in online algorithms with predictions and smoothed online classification, which go beyond a worse-case analysis by using machine-learned predictions and distributional assumptions respectively.