K-means Classifier Shown to Be a GMM Special Case, Enabling OOD Detection

A researcher shows that K-means classifiers are a special case of GMMs, allowing likelihood-based OOD detection, and further integrates the K-means nonlinear classifier into an alternating minimization training framework.

2026-11-11 ~ 2026-11-11 · 2 related posts

• K-means-based output classifier still trainable via alternating minimization, with a KL-constrained E-step — caglar_ee · 2026-11-11
• K-means classifier is a GMM special case: likelihood enables OOD detection — caglar_ee · 2026-11-11