arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1702.02670stat.MLmath.STstat.TH
收藏

随机邻居嵌入分离良好的聚类

Stochastic Neighbor Embedding separates well-separated clusters

Uri Shaham, Stefan Steinerberger

摘要:

随机邻域嵌入及其变体是广泛使用的降维技术-尽管它们很受欢迎,但没有理论结果。我们证明了从高维到任何欧几里得空间R^d的良好分离的聚类的最佳SNE嵌入能够以定量的方式成功地分离聚类。结果也适用于一个更大的家庭的方法,包括一个变种的t-SNE。

英文摘要:

Stochastic Neighbor Embedding and its variants are widely used dimensionality reduction techniques -- despite their popularity, no theoretical results are known. We prove that the optimal SNE embedding of well-separated clusters from high dimensions to any Euclidean space R^d manages to successfully separate the clusters in a quantitative way. The result also applies to a larger family of methods including a variant of t-SNE.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑