Contrastive learning unifies $t$-SNE and UMAP. (arXiv:2206.01816v1 [cs.LG])

Neighbor embedding methods $t$-SNE and UMAP are the de facto standard for
visualizing high-dimensional datasets. They appear to use very different loss
functions with different motivations, and the exact relationship between them
has been unclear. Here we show that UMAP is effectively negative sampling
applied to the $t$-SNE loss function. We explain the difference between
negative sampling and noise-contrastive estimation (NCE), which has been used
to optimize $t$-SNE under the name NCVis. We prove that, unlike NCE, negative
sampling learns …

arxiv learning umap