The story’s score and claims

Comments on No. 42 of 95 · e-means.github.io · edition of Tuesday 10 November 2026

E-means: 10x faster k-means at equal loss

Summary

E-means is a mini-batch k-means method that replaces Lloyd's full-batch updates with exponential moving averages of centroid sums and counts, plus cosine-scheduled step sizes, dead-cluster revival and a finalization stage.

Comments

Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules