Table of Contents
Fetching ...

Machine Learning vs. SED-Fitting: A Comparative Analysis of Accuracy in Stellar Mass Estimation

Vahid Asadi, Akram Hasani Zonoozi, Hosein Haghi

TL;DR

This study tackles the persistent biases and computational bottlenecks in photometric stellar-mass estimation by performing a controlled comparison between a machine-learning approach (Pt-SNE) and a traditional SED-fitting method (LePhare). Using a Horizon-AGN COSMOS-like mock with known true masses, Pt-SNE is trained on noise-injected BC03 templates and evaluated against LePhare across 12 photometric bands, including analyses with band-reductions and varying signal-to-noise. Pt-SNE achieves substantially lower bias ($\text{Bias}=0.029$ dex) and a smaller RMS ($\sigma_F=0.169$ dex) than LePhare ($\text{Bias}=0.286$ dex, $\sigma_F=0.306$ dex), while also exhibiting superior robustness (lower $\sigma_{NMAD}$ and OLF) and requiring far fewer bands (12 vs 26) for comparable or better accuracy. Moreover, Pt-SNE is dramatically faster—approximately $3.2\times10^3$ times faster on large datasets—highlighting its suitability for scalable analyses in upcoming large-area surveys. The study also reveals limitations related to training-data representativeness, showing that extrapolation beyond the training domain can induce biases, thereby underscoring the need for diverse training sets and possibly broader template libraries for real-world applications.

Abstract

Traditional spectral energy distribution (SED)-fitting methods for stellar mass estimation face persistent challenges including systematic biases and computational constraints. We present a controlled comparison of machine learning (ML) and SED-fitting methods, assessing their accuracy, robustness, and computational efficiency. Using a sample of COSMOS-like galaxies from the Horizon-AGN simulation as a benchmark with known true masses, we evaluate the Parametric t-SNE (Pt-SNE) algorithm -- trained on noise-injected BC03 models -- against the established SED-fitting code LePhare. Our results demonstrate that Pt-SNE achieves superior accuracy, with a root-mean-square error (sigma_F) of 0.169 dex compared to LePhare's 0.306 dex. Crucially, Pt-SNE exhibits significantly lower bias (0.029 dex) compared to LePhare (0.286 dex). Pt-SNE also shows greater robustness across all stellar mass ranges, particularly for low-mass galaxies (10^9 to 10^10 solar masses), where it reduces errors by 47-53 %. Even when restricted to only six optical bands, Pt-SNE outperforms LePhare using all 26 available photometric bands, underscoring its superior informational efficiency. Computationally, Pt-SNE processes large datasets approximately 3.2 x 10^3 times faster than LePhare. These findings highlight the fundamental advantages of ML methods for stellar mass estimation, demonstrating their potential to deliver more accurate, stable, and scalable measurements for large-scale galaxy surveys.

Machine Learning vs. SED-Fitting: A Comparative Analysis of Accuracy in Stellar Mass Estimation

TL;DR

This study tackles the persistent biases and computational bottlenecks in photometric stellar-mass estimation by performing a controlled comparison between a machine-learning approach (Pt-SNE) and a traditional SED-fitting method (LePhare). Using a Horizon-AGN COSMOS-like mock with known true masses, Pt-SNE is trained on noise-injected BC03 templates and evaluated against LePhare across 12 photometric bands, including analyses with band-reductions and varying signal-to-noise. Pt-SNE achieves substantially lower bias ( dex) and a smaller RMS ( dex) than LePhare ( dex, dex), while also exhibiting superior robustness (lower and OLF) and requiring far fewer bands (12 vs 26) for comparable or better accuracy. Moreover, Pt-SNE is dramatically faster—approximately times faster on large datasets—highlighting its suitability for scalable analyses in upcoming large-area surveys. The study also reveals limitations related to training-data representativeness, showing that extrapolation beyond the training domain can induce biases, thereby underscoring the need for diverse training sets and possibly broader template libraries for real-world applications.

Abstract

Traditional spectral energy distribution (SED)-fitting methods for stellar mass estimation face persistent challenges including systematic biases and computational constraints. We present a controlled comparison of machine learning (ML) and SED-fitting methods, assessing their accuracy, robustness, and computational efficiency. Using a sample of COSMOS-like galaxies from the Horizon-AGN simulation as a benchmark with known true masses, we evaluate the Parametric t-SNE (Pt-SNE) algorithm -- trained on noise-injected BC03 models -- against the established SED-fitting code LePhare. Our results demonstrate that Pt-SNE achieves superior accuracy, with a root-mean-square error (sigma_F) of 0.169 dex compared to LePhare's 0.306 dex. Crucially, Pt-SNE exhibits significantly lower bias (0.029 dex) compared to LePhare (0.286 dex). Pt-SNE also shows greater robustness across all stellar mass ranges, particularly for low-mass galaxies (10^9 to 10^10 solar masses), where it reduces errors by 47-53 %. Even when restricted to only six optical bands, Pt-SNE outperforms LePhare using all 26 available photometric bands, underscoring its superior informational efficiency. Computationally, Pt-SNE processes large datasets approximately 3.2 x 10^3 times faster than LePhare. These findings highlight the fundamental advantages of ML methods for stellar mass estimation, demonstrating their potential to deliver more accurate, stable, and scalable measurements for large-scale galaxy surveys.
Paper Structure (22 sections, 6 equations, 9 figures, 2 tables)

This paper contains 22 sections, 6 equations, 9 figures, 2 tables.

Figures (9)

  • Figure 1: Transmission curves for the photometric bands used for machine learning method.
  • Figure 2: Manifold of Pt-SNE (black points), with the mapped COSMOS-like mock galaxies (colored points). Galaxy colors correspond to the estimated stellar masses obtained with Pt-SNE. The dimensions of the 2D plot (D1 and D2) are arbitrary labels and do not carry any physical significance.
  • Figure 3: The residual distributions show a significant systematic bias in LePhare (red) versus the unbiased Pt-SNE results (black). LePhare's tighter distribution indicates precision but poor accuracy, while Pt-SNE provides more accurate mass estimates. The dashed line marks perfect agreement.
  • Figure 4: Comparison of the stellar masses of 91,261 COSMOS-like galaxies, derived using Pt-SNE ($M_{Pt-SNE}$) and LePhare ($M_{LePhare}$), with those obtained from simulation ($M_{Sim}$). $\sigma_F$, $\sigma_{STD}$ and Bias are root-mean-square, standard deviation and mean offset respectively.
  • Figure 5: Error trends across simulation-based mass bins ($\log(M/M_{\odot})$) for SNR $> 1.5$ (low-SNR) and SNR $> 150$ (high-SNR). Metrics shown are the root-mean-square error ($\sigma_F$, Equation \ref{['eq:rms']}) and bias (Bias, Equation \ref{['eq:bias']})
  • ...and 4 more figures