Stochastic Neighbor Embedding separates well-separated clusters

2017/02/09 by Uri Shaham, Stefan Steinerberger, Shaham, Uri +1 · 2 citations
Computer Science · Mathematics · #FOS: Computer and information sciences #FOS: Mathematics #Face and Expression Recognition #Gaussian Processes and Bayesian Inference #Machine Learning (stat.ML) #Statistical Methods and Inference #Statistics Theory (math.ST)

paper · pdf · doi:10.48550/arxiv.1702.02670

openalex publication_date 2017/02/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

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.

Citations

• Distributing many points on a sphere

Cited by

• t-SNE Exaggerates Clusters, Provably
• Compressibility Barriers to Neighborhood-Preserving Data Visualizations

Related

• Clustering with t-SNE, provably
• T-SNE Is Not Optimized to Reveal Clusters in Data
• Kernel t-distributed stochastic neighbor embedding
• ENS-t-SNE: Embedding Neighborhoods Simultaneously t-SNE
• EmbedOR: Provable Cluster-Preserving Visualizations with Curvature-Based Stochastic Neighbor Embeddings