EasyDeepLearn

PCA vs t-SNE vs UMAP — when do you use each?

medium

Answer

  • PCA is a linear, deterministic projection that preserves global variance — use it for compression, denoising, or as input to another model. t-SNE and UMAP are nonlinear and preserve local neighborhoods — use them for 2D/3D visualization, not for downstream modeling.
  • UMAP is faster than t-SNE and preserves more global structure.
  • Never interpret t-SNE distances between clusters as meaningful.
Check yourself — multiple choice
  • t-SNE is the best choice for feeding into another model
  • PCA is nonlinear; t-SNE is linear
  • PCA for downstream ML, t-SNE/UMAP for visualization
  • UMAP is slower than t-SNE

PCA is a linear tool used for modeling; t-SNE/UMAP are visualization tools.

#dimensionality-reduction#visualization

Practise Unsupervised Learning

214 interview questions in this topic.

Related questions