Accelerating t-SNE using Tree-Based Algorithms
AUTHOR(S)
van der Maaten, Laurens
PUB. DATE
October 2014
SOURCE
Journal of Machine Learning Research;2014, Vol. 15, p3221
SOURCE TYPE
Academic Journal
DOC. TYPE
Article
ABSTRACT
The paper investigates the acceleration of t-SNE--an embedding technique that is commonly used for the visualization of high-dimensional data in scatter plots--using two tree-based algorithms. In particular, the paper develops variants of the Barnes-Hut algorithm and of the dual-tree algorithm that approximate the gradient used for learning t-SNE em-beddings in O(N log N). Our experiments show that the resulting algorithms substantially accelerate t-SNE, and that they make it possible to learn embeddings of data sets with millions of objects. Somewhat counterintuitively, the Barnes-Hut variant of t-SNE appears to outperform the dual-tree variant.
本文使用两种基于树的算法研究了t-SNE的加速特性,t-SNE是一种常用的嵌入技术,用于散点图中高维数据的可视化。特别是,本文开发的变体Barnes-Hut算法和近似梯度的dual-tree算法用于学习t-SNE em-beddings在O (N log N)。
实验表明,生成的算法大大加快t-SNE,,他们可以学习与数以百万计的对象嵌入的数据集。
有点反直觉的是,t-SNE的Barnes-Hut变体似乎比双树变体表现得更好。