Wiki/Topics/AI / ML/Unsupervised/Dimensionality Reduction/UMAP

UMAP

dimensionality-reductionumaptopologyunsupervised2026-04-08

I Use This When...

<!-- Practical use case -->

History

Leland McInnes, John Healy (2018). Based on topological data analysis. Faster than t-SNE, better at preserving global structure.

Why It Exists

t-SNE is slow on large datasets and doesn't preserve global structure well. UMAP is faster, scales better, and maintains both local and global relationships.

How It Works

Visual Intuition

<!-- 3B1B-style animation description -->

Step by Step

<!-- Algorithm walkthrough -->

Code

# Implementation sketch

The Math Inside

Builds a weighted k-nearest-neighbor graph in high-D. Optimizes a low-D layout that preserves the topological structure (fuzzy simplicial set).

Math Prerequisites

<!-- Links to math wiki -->
  • t-SNE — Predecessor
  • PCA — Linear baseline

Linked from