Re:E-means: 10x faster k-means at equal loss
by Thomashuet on Tuesday November 10, 2026 @03:38PM
(#50445990)
I understand that e-means does the same thing as k-means but faster, that's great, congratulations on that. I'm just saying that the example you've chosen to showcase this result is terrible. The goal of a clustering algorithm is to discover the structure of the data and in your example the structure is simply not there (to be precise, there is only a bit of structure but you're trying to find a lot more of it than there is). I think it would be much more convincing if you showed an example where the number of clusters you're looking for actually matches the number of clusters in the data.
The goal of e-means is to be a drop-in replacement for k-means. It solves the exact same problem, with the exact same inputs and delivers better centroids and quality in a fraction of the time. So I'm not sure what we get wrong by working on a problem that hundreds of papers also worked on. Is the whole k-means literature getting it wrong then ?
Picking the k is a separate problem, on which also a body of literature has worked too. e-means is a drop-in replacement for k-means so you can use any method you would use for choosing the k but with e-means instead of k-means.
e-means performs better on a very wide array of datasets, definitely not degenerate cases, don't hesitate to check out our paper at https://arxiv.org/abs/2611.09501.
If you have any question or objection, I'll be happy to answer.