Learning Selection Cuts With Gradients
Mike Hance, Juan Robles · 2025
Read Learning Selection Cuts With Gradients by Mike Hance, Juan Robles (2025) 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-2502-08615.txt.
Description
Mike Hance and Juan Robles teach selection cuts with gradient methods in 2025. Particle physics searches filter events by thresholds on measured traits. Hand tuning those cuts is slow, and many teams skip it for black box classifiers. The authors recast each cut as a learned value inside a simple network.
Custom loss rules steer the cuts toward a target share of kept signal events. Extra terms keep the cuts smooth as conditions change. The method was tried on a search for supersymmetry and matched common tools well. Code is shared in a public repository so other teams can use the approach.
Catalog record
| CREATOR | Mike Hance, Juan Robles |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-08615