Optimal Demixing of Nonparametric Densities
Jianqing Fan [email protected], Zheng Tracy Ke [email protected], Zhaoyang Shi [email protected] · 2026
Read Optimal Demixing of Nonparametric Densities by Jianqing Fan [email protected], Zheng Tracy Ke [email protected], Zhaoyang Shi [email protected] (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-2603-27457.txt.
Introduction
Fan and co-authors tackle unmixing blended data distributions. Each group of samples mixes the same hidden densities in new shares. The task is to recover each hidden density. Their method adapts kernel density estimation with group weights. Weights come from topic-style fits on histogram data. A U-statistic step removes bias from those weights. Error rates match a proven lower bound, so the method is optimal. The problem extends topic modeling to smooth data. It also applies to word embedding models.
Catalog record
| CREATOR | Jianqing Fan [email protected], Zheng Tracy Ke [email protected], Zhaoyang Shi [email protected] |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 197353 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.27457