Improved Strongly Adaptive Online Learning using Coin Betting
This work addresses the challenge of adaptive online learning for applications like expert advice and metric learning, representing an incremental improvement over existing methods.
The paper tackles the problem of online learning in changing environments by introducing a new parameter-free algorithm, achieving a strongly adaptive regret bound that is at least a factor of $\sqrt{\log(T)}$ better than comparable algorithms, with empirical results showing outperformance in expert advice and metric learning scenarios.
This paper describes a new parameter-free online learning algorithm for changing environments. In comparing against algorithms with the same time complexity as ours, we obtain a strongly adaptive regret bound that is a factor of at least $\sqrt{\log(T)}$ better, where $T$ is the time horizon. Empirical results show that our algorithm outperforms state-of-the-art methods in learning with expert advice and metric learning scenarios.