EasyDeepLearn

GMM covariance types — which do you pick?

hard

Answer

  • Full: each cluster has its own general Σj{\Sigma}_{j} — most flexible, most parameters (p2)(p^{2}).
  • Tied: all clusters share one Σ.
  • Diagonal: axis-aligned ellipsoids (p params per cluster).
  • Spherical: Σ  =  σ2{\Sigma}\; = \;{\sigma}^{2} I (1 param) — closest to k-means.
  • Rule: use full for small p, tied or diagonal for high p to avoid overfitting.
  • Cross-validate log-likelihood or BIC to choose.
Check yourself — multiple choice
  • Only full
  • Full (flexible,  p2)  /  tied(\mathrm{flexible}, \;p^{2})\; / \;\mathrm{tied} (shared) / diagonal (axis-aligned) / spherical (~k-means). Full for small p, diagonal/tied for high p; select by BIC
  • Random
  • Same as k-means

GMM covariance: full / tied / diagonal / spherical; trade flexibility vs params.

#clustering#density

Practise Unsupervised Learning

214 interview questions in this topic.

Related questions