EasyDeepLearn

t-SNE perplexity — what does it control?

medium

Answer

  • Perplexity ≈ effective number of neighbors each point 'considers' when building the conditional similarity distribution.
  • Typical range: 5-50.
  • Low perplexity → local, may fragment groups.
  • High → global, may merge distinct clusters.
  • Rule of thumb: try 5, 30, 50 and inspect.
  • Don't over-tune to make plots look better — the map should be robust across a range.
Check yourself — multiple choice
  • Random
  • Effective #neighbors per point; low → local/fragmented; high → global/merged; try 5, 30, 50 — plot should be robust across range
  • Same as k
  • Not real

Perplexity: effective neighbor count; typical 5-50; try multiple values.

#dimensionality-reduction#visualization

Practise Unsupervised Learning

214 interview questions in this topic.

Related questions