TriMap: Large-scale Dimensionality Reduction Using Triplets

TL;DR

TriMap is a scalable dimensionality reduction method using triplet constraints that better preserves global data structure than existing techniques like t-SNE, LargeVis, and UMAP, with strong empirical performance.

Contribution

The paper introduces TriMap, a novel large-scale dimensionality reduction technique based on triplet constraints that improves global structure preservation and scalability.

Findings

TriMap outperforms t-SNE, LargeVis, and UMAP in embedding quality.

TriMap scales to millions of points efficiently.

TriMap has faster runtime and better global structure preservation.

Abstract

We introduce "TriMap"; a dimensionality reduction technique based on triplet constraints, which preserves the global structure of the data better than the other commonly used methods such as t-SNE, LargeVis, and UMAP. To quantify the global accuracy of the embedding, we introduce a score that roughly reflects the relative placement of the clusters rather than the individual points. We empirically show the excellent performance of TriMap on a large variety of datasets in terms of the quality of the embedding as well as the runtime. On our performance benchmarks, TriMap easily scales to millions of points without depleting the memory and clearly outperforms t-SNE, LargeVis, and UMAP in terms of runtime.