Fast k k -means Seeding Under The Manifold Hypothesis
Poojan Shah, Shashwat Agrawal, Ragesh Jaiswal · 2026
Read Fast k k -means Seeding Under The Manifold Hypothesis by Poojan Shah, Shashwat Agrawal, Ragesh Jaiswal (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-01104.txt.
Description
Poojan Shah and co-authors model run-of-the-mill k-means cases, not worst cases that rarely show up. Past proofs lean on claims about best answers that are hard to test in real data. Their manifold hunch says real data in high space hugs a low-space sheet, and they pin down shape traits with set scaling laws from best-quantization math. Those traits drive Qkmeans, a fast seed plan with fresh speed-versus-score trades, backed by a broad test sweep.
Catalog record
| CREATOR | Poojan Shah, Shashwat Agrawal, Ragesh Jaiswal |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 73670 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.01104