Optimal and Adaptive Algorithms for Online Boosting
This work addresses the challenge of online boosting for machine learning practitioners, offering both optimal and adaptive solutions, though it is incremental as it builds on existing boosting concepts.
The authors tackled the problem of converting weak online learners into strong ones by developing two online boosting algorithms, one optimal but non-adaptive and another adaptive but suboptimal, with the optimal algorithm proven to require minimal weak learners and sample complexity for specified accuracy.
We study online boosting, the task of converting any weak online learner into a strong online learner. Based on a novel and natural definition of weak online learnability, we develop two online boosting algorithms. The first algorithm is an online version of boost-by-majority. By proving a matching lower bound, we show that this algorithm is essentially optimal in terms of the number of weak learners and the sample complexity needed to achieve a specified accuracy. This optimal algorithm is not adaptive however. Using tools from online loss minimization, we derive an adaptive online boosting algorithm that is also parameter-free, but not optimal. Both algorithms work with base learners that can handle example importance weights directly, as well as by rejection sampling examples with probability defined by the booster. Results are complemented with an extensive experimental study.