Photometric Redshift PDFs via Neural Network Classification for DESI Legacy Imaging Surveys and Pan-STARRS
Da-Chuan Tian, Zhong-Lue Wen, Jun-Qing Xia · 2026
Read Photometric Redshift PDFs via Neural Network Classification for DESI Legacy Imaging Surveys and Pan-STARRS by Da-Chuan Tian, Zhong-Lue Wen, Jun-Qing Xia (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-2602-01548.txt.
Description
Da-Chuan Tian and co-authors estimate photometric redshifts by classification, giving well-calibrated probability curves. Their network splits redshift space into ordered bins, optimizing a continuous ranked probability score, keeping redshift's ordinal nature, giving uncertainty. Unlike fits giving single point values, classification captures multi-modal posteriors born of color-redshift degeneracies. On a record spectral sample from DESI and SDSS, it beats Random Forest, XGBoost, and neural regression on DESI Legacy and Pan-STARRS. Deep optical photometry plus mid-infrared coverage proves essential for high precision across the range.
Catalog record
| CREATOR | Da-Chuan Tian, Zhong-Lue Wen, Jun-Qing Xia |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59895 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.01548