嵌套小批量K-Means

Feb, 2016

Turbocharging Mini-Batch K-Means

James Newling, François Fleuret

TL;DR通过使用Elkan（2003）的距离界限方法来加速Sculley（2010）的Mini-Batch K-Means算法，提出了一种新算法。使用嵌套的Mini-Batches提出了两个困难。实验表明，得到的嵌套Mini-Batch算法非常有效，通常比标准Mini-Batch算法早100次到达经验最小值的1%。

Abstract

We propose an accelerated mini-batch k-means algorithm which combines three key improvements. The first is a modified center update which results in convergence to a local minimum in fewer iterations. The second