Abstract:We present PopSED, a framework for the population-level inference of galaxy properties from photometric data. Unlike the traditional approach of first analyzing individual galaxies and then combining the results to determine the physical properties of the entire galaxy population, we directly make the population distribution the inference objective. We train normalizing flows to approximate the population distribution by minimizing the Wasserstein distance between the synthetic photometry of the galaxy population and the observed data. We validate our method using mock observations and apply it to galaxies from the GAMA survey. PopSED reliably recovers the redshift and stellar mass distribution of $10^{5}$ galaxies using broadband photometry within $<1$ GPU hr, being $10^{5-6}$ times faster than the traditional spectral energy distribution modeling method. From the population posterior, we also recover the star-forming main sequence for GAMA galaxies at $z<0.1$. With the unprecedented number of galaxies in upcoming surveys, our method offers an efficient tool for studying galaxy evolution and deriving redshift distributions for cosmological analyses.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to improve the efficiency and accuracy of inferring the properties of galaxy populations from photometric data. Specifically, traditional methods usually first analyze the spectral energy distribution (SED) of individual galaxies and then aggregate the results to determine the physical properties of the entire galaxy population. This method is very time - consuming and computationally expensive, especially for data sets of a large number of galaxies. To solve this problem, the authors propose a new framework, PopSED, which directly takes the distribution of the galaxy population as the inference target instead of first fitting individual galaxies. By training normalizing flows to approximate the population distribution and minimizing the Wasserstein distance between the synthetic photometry and the observed data, PopSED can reliably recover the redshift and stellar mass distributions of 100,000 galaxies in less than 1 GPU - hour, which is 100,000 to 1,000,000 times faster than traditional methods. In addition, PopSED can also recover the star - formation main - sequence relationship of GAMA galaxies at z < 0.1 from the population posterior. With the dramatic increase in the number of galaxies in future surveys, PopSED provides an efficient tool to study galaxy evolution and derive redshift distributions for cosmological analysis. In summary, the main contributions of this paper are: - Proposing an efficient and robust method, PopSED, for inferring the properties of galaxy populations from broadband photometric data. - Using normalizing flows and Wasserstein distance to optimize the approximation of the population distribution, avoiding fitting individual galaxies one by one. - Significantly improving computational efficiency when dealing with large - scale galaxy data, making it possible to study the entire galaxy population. In terms of formulas, the Wasserstein distance is used in the paper to measure the difference between the observed data and the synthetic data, which is defined as follows: \[ W_2(\{\hat{X}_j\}, \{X_i\}) = \left( \inf_{\gamma \in \Pi(\{\hat{X}_j\}, \{X_i\})} \sum_{i,j} \| \hat{X}_j - X_i \|^2 \gamma_{ij} \right)^{1/2} \] where $\Pi(\{\hat{X}_j\}, \{X_i\})$ represents the set of all joint distributions from $\{\hat{X}_j\}$ to $\{X_i\}$, and $\gamma_{ij}$ is an element in these joint distributions. Hopefully, this information will help you better understand the core problem of this paper and its solution. If you have more questions or need further explanation, please feel free to let me know.

PopSED: Population-Level Inference for Galaxy Properties from Broadband Photometry with Neural Density Estimation

pop-cosmos: Scaleable Inference of Galaxy Properties and Redshifts with a Data-driven Population Model

From Halos to Galaxies. X: Decoding Galaxy SEDs with Physical Priors and Accurate Star Formation History Reconstruction

BayeSED-GALAXIES I. Performance test for simultaneous photometric redshift and stellar population parameter estimation of galaxies in the CSST wide-field multiband imaging survey

pop-cosmos: A comprehensive picture of the galaxy population from COSMOS data

Galaxy spectroscopy without spectra: Galaxy properties from photometric images with conditional diffusion models

Decomposition of stellar populations in CosmoDC2 galaxies using SCARLET and Deep Learning

Spatially Resolved Galaxy-Dust Modeling with Coupled Data-Driven Priors

Stellar Populations With Optical Spectra: Deep Learning vs. Popular Spectrum Fitting Codes

RadioSED I: Bayesian inference of radio SEDs from inhomogeneous surveys

Starduster: A multi-wavelength SED model based on radiative transfer simulations and deep learning

PixelPop: High Resolution Nonparameteric Inference of Gravitational-Wave Populations in Multiple Dimensions

Can We Constrain Galaxy Geometry Parameters Using Spatially Integrated SED Fitting?

Photometric redshift estimation via deep learning

GalSBI: Phenomenological galaxy population model for cosmology using simulation-based inference

Application of Manifold Learning to Selection of Different Galaxy Populations and Scaling Relation Analysis

Silkscreen: Direct Measurements of Galaxy Distances from Survey Image Cutouts