Deep Learning Improves Photometric Redshifts in All Regions of Color Space

TL;DR

This paper demonstrates that deep learning methods significantly improve photometric redshift accuracy across all galaxy color regions, reducing bias and scatter compared to classical machine learning techniques, especially in complex color spaces.

Contribution

The study shows that deep learning photo-$z$ methods outperform classical ML in all color space regions and exhibit less color-dependent bias, highlighting their robustness for upcoming cosmological surveys.

Findings

Deep learning reduces photo-$z$ scatter across all color regions.

Classical ML suffers from color-dependent attenuation bias.

Deep learning effectively minimizes systematic biases in redshift estimation.

Abstract

Photometric redshifts (photo-$z$'s) are crucial for the cosmology, galaxy evolution, and transient science drivers of next-generation imaging facilities like the Euclid Mission, the Rubin Observatory, and the Nancy Grace Roman Space Telescope. Previous work has shown that image-based deep learning photo-$z$ methods produce smaller scatter than photometry-based classical machine learning (ML) methods on the Sloan Digital Sky Survey (SDSS) Main Galaxy Sample, a testbed photo-$z$ dataset. However, global assessments can obscure local trends. To explore this possibility, we used a self-organizing map (SOM) to cluster SDSS galaxies based on their $ugriz$ colors. Deep learning methods achieve lower photo-$z$ scatter than classical ML methods for all SOM cells. The fractional reduction in scatter is roughly constant across most of color space with the exception of the most bulge-dominated and reddest cells where it is smaller in magnitude. Interestingly, classical ML photo-$z$'s suffer from a significant color-dependent attenuation bias, where photo-$z$'s for galaxies within a SOM cell are systematically biased towards the cell's mean spectroscopic redshift and away from extreme values, which is not readily apparent when all objects are considered. In contrast, deep learning photo-$z$'s suffer from very little color-dependent attenuation bias. The increased attenuation bias for classical ML photo-$z$ methods is the primary reason why they exhibit larger scatter than deep learning methods. This difference can be explained by the deep learning methods weighting redshift information from the individual pixels of a galaxy image more optimally than integrated photometry.