UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

TL;DR

UMAP is a scalable manifold learning algorithm for dimension reduction that preserves global data structure better than t-SNE and is applicable to high-dimensional data in machine learning.

Contribution

It introduces a new dimension reduction technique based on Riemannian geometry and algebraic topology, offering improved scalability and global structure preservation.

Findings

UMAP is competitive with t-SNE in visualization quality.

UMAP preserves more global structure than t-SNE.

UMAP has superior runtime performance.

Abstract

UMAP (Uniform Manifold Approximation and Projection) is a novel manifold learning technique for dimension reduction. UMAP is constructed from a theoretical framework based in Riemannian geometry and algebraic topology. The result is a practical scalable algorithm that applies to real world data. The UMAP algorithm is competitive with t-SNE for visualization quality, and arguably preserves more of the global structure with superior run time performance. Furthermore, UMAP has no computational restrictions on embedding dimension, making it viable as a general purpose dimension reduction technique for machine learning.