Transferability of Photometric Redshifts Determined using Machine Learning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916339521683456 |
|---|---|
| author | Janiurek, Lara Hendry, Martin A. Speirits, Fiona C. |
| author_facet | Janiurek, Lara Hendry, Martin A. Speirits, Fiona C. |
| contents | In this work the random forest algorithm GALPRO is implemented to generate photometric redshift posteriors, and its performance when trained and then applied to data from another survey is investigated. The algorithm is initially calibrated using a truth dataset compiled from the DESI Legacy survey. We find that the testing and training datasets must have very similar redshift distributions, with the range of their photometric data overlapping by at least 90% in the appropriate photometric bands in order for the training data to be applicable to the testing data. Then GALPRO is again trained using the DESI dataset and then applied to a sample drawn from the PanSTARRS survey, to explore whether GALPRO can be first trained using a trusted dataset and then applied to an entirely new survey, albeit one that uses a different magnitude system for its photometric bands, thus requiring careful conversion of the measured magnitudes for the new survey before GALPRO can be applied. The results of this further test indicate that GALPRO does not produce accurate photometric redshift posteriors for the new survey, even where the distribution of redshifts for the two datasets overlaps by over 90%. Hence, we conclude that the photometric redshifts generated by GALPRO are not suitable for generating estimates of photometric redshifts and their posterior distribution functions when applied to an entirely new survey, particularly one that uses a different magnitude system. However, our results demonstrate that GALPRO is a useful tool for inferring photometric redshift estimates in the case where a spectroscopic galaxy survey is nearly complete, but is missing some spectroscopic redshift values. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_20670 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Transferability of Photometric Redshifts Determined using Machine Learning Janiurek, Lara Hendry, Martin A. Speirits, Fiona C. Astrophysics of Galaxies 85 In this work the random forest algorithm GALPRO is implemented to generate photometric redshift posteriors, and its performance when trained and then applied to data from another survey is investigated. The algorithm is initially calibrated using a truth dataset compiled from the DESI Legacy survey. We find that the testing and training datasets must have very similar redshift distributions, with the range of their photometric data overlapping by at least 90% in the appropriate photometric bands in order for the training data to be applicable to the testing data. Then GALPRO is again trained using the DESI dataset and then applied to a sample drawn from the PanSTARRS survey, to explore whether GALPRO can be first trained using a trusted dataset and then applied to an entirely new survey, albeit one that uses a different magnitude system for its photometric bands, thus requiring careful conversion of the measured magnitudes for the new survey before GALPRO can be applied. The results of this further test indicate that GALPRO does not produce accurate photometric redshift posteriors for the new survey, even where the distribution of redshifts for the two datasets overlaps by over 90%. Hence, we conclude that the photometric redshifts generated by GALPRO are not suitable for generating estimates of photometric redshifts and their posterior distribution functions when applied to an entirely new survey, particularly one that uses a different magnitude system. However, our results demonstrate that GALPRO is a useful tool for inferring photometric redshift estimates in the case where a spectroscopic galaxy survey is nearly complete, but is missing some spectroscopic redshift values. |
| title | Transferability of Photometric Redshifts Determined using Machine Learning |
| topic | Astrophysics of Galaxies 85 |
| url | https://arxiv.org/abs/2407.20670 |