Metraipolis › archive › dc_af9754cc6734
e-means: outperforming k-means (in one-tenth the epochs) — arXiv:2611.09501
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.09501
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-09T22:35:06.336Z
- SHA-256 of the archived text
e3ee1d76e8bf2e3d6df0bfb4729ef5c4c619e30e0e236ffb8edf1be05aa7a20d- Stable URL of this page
https://metraipolis.com/documento/dc_af9754cc6734
How to cite
e-means: outperforming k-means (in one-tenth the epochs) — arXiv:2611.09501 — arXiv (en), https://arxiv.org/abs/2611.09501 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_af9754cc6734 (Metraipolis archive dc_af9754cc6734, sha256 e3ee1d76e8bf2e3d6df0bfb4729ef5c4c619e30e0e236ffb8edf1be05aa7a20d)
BibTeX
@misc{dcaf9754cc6734,
title = {e-means: outperforming k-means (in one-tenth the epochs) — arXiv:2611.09501},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.09501},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_af9754cc6734; sha256 e3ee1d76e8bf2e3d6df0bfb4729ef5c4c619e30e0e236ffb8edf1be05aa7a20d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09501. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Lloyd's k-means algorithm is a long-running standard for clustering, with applications in data analysis, retrieval, quantization and data curation. In this paper, we introduce an exponential moving average optimization procedure for k-means with mini-batch updates, powered by modern deep learning machinery and new insights. We also establish scaling laws on how decay rate and batch size impact final performance. This gives us a formulaic rule to fix them for multiple problem sizes and embedding datasets, without the need for any hyperparameter sweep. The resulting clustering algorithm, coined e-means, is much faster than Lloyd k-means and converges to lower losses, frequently even beating k-means++. On a wide array of embeddings datasets, it beats the loss of Lloyd k-means at 25 epochs in 1.4 to 2.6 epochs, which translates to 10-45x GPU wall-clock speedups compared to state-of-the-art k-means implementations. As an example, e-means partitions DINO10B, a dataset of 10 billion DINO embeddings, into 16 million clusters in only 2.5 epochs, reaching a loss below Lloyd's at 25 epochs. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.09501 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-09T22:30:47.335Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).