BriefGPT.xyz
Feb, 2015
在线 boosting 的最优与自适应算法
Optimal and Adaptive Algorithms for Online Boosting
HTML
PDF
Alina Beygelzimer, Satyen Kale, Haipeng Luo
TL;DR
研究在线提升的两种算法：boost-by-majority和自适应在线提升算法，证明其在可接受精度下基本上是最优的。两种算法都能处理样本的重要性权重，并且使用了在线损失最小化技术。
Abstract
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
→