Leave-One-Out Neighborhood Smoothing for Graphons: Berry-Esseen Bounds, Confidence Intervals, and Honest Tuning
Behzad Aalipur, Rachel Kilby · 2026
Read Leave-One-Out Neighborhood Smoothing for Graphons: Berry-Esseen Bounds, Confidence Intervals, and Honest Tuning by Behzad Aalipur, Rachel Kilby (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-2602-02319.txt.
Description
Behzad Aalipur and Rachel Kilby study uncertainty quantification in graphon models, where gauging guarantees lack tractable theory for edge probabilities. They present a leave-one-out neighborhood smoother building neighborhoods without target-column edges, so averaged notes become independent Bernoulli variables with differing probabilities. That separation yields concentration bounds and a Berry-Esseen-type normal-approximation bound for every fixed target pair, with a pair-independent numeric constant. That leave-one-out structure gives an edge-wise no-leakage checks identity, since the held-out edge never builds its predictor.
Catalog record
| CREATOR | Behzad Aalipur, Rachel Kilby |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 63383 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.02319