Using Active Learning to Improve Quasar Identification for the DESI Spectra Processing Pipeline
H. Zou · 2025
Read Using Active Learning to Improve Quasar Identification for the DESI Spectra Processing Pipeline by H. Zou (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-2505-01596.txt.
Description
Dylan Green and a large survey team teach a quasar finder with active learning in a 2025 paper. A big sky survey classifies millions of spectra automatically. Its quasar network was trained on an older survey's data and needed refreshing with new data. Labeling spectra by hand is slow, so the team let an algorithm choose which spectra were worth labeling.
Their active learning loop picks the most informative examples for training. A self-organizing map step rejects oddball spectra so the labeled set still mirrors the true quasar population. The retrained network classifies quasars more accurately on the new survey's data. The method shows how to spend expert labeling time where it buys the most accuracy.
Catalog record
| CREATOR | H. Zou |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-01596