[论文解读] A Demonstration of a Neural Network as a Bridge Between Galaxy Simulations and Surveys

E. Elson|arXiv (Cornell University)|Feb 6, 2026

Galaxies: Formation, Evolution, Phenomena被引用 0

一句话总结

一个简单、基于仿真训练的神经网络可以从宽带光度预测GAMA星系的恒星质量，在约3.5 dex的范围内与SED推导的质量相比，散射约0.131 dex；对于缺少SED质量的星系，保守不确定度约为0.18 dex。

ABSTRACT

This paper demonstrates that the stellar masses of galaxies in the Galaxy and Mass Assembly (GAMA) survey, originally derived via stellar population synthesis modelling, can be accurately predicted using only their absolute magnitudes and colour indices. A central contribution of this work is the demonstration that this long-standing inference problem can be solved using an exceptionally simple machine-learning model: a fully connected, feed-forward artificial neural network with a single hidden layer. The network is trained exclusively on synthetic galaxies generated by the SHARK semi-analytic model and is shown to transfer effectively to real observations. Across nearly 3.5 dex in stellar mass, the predicted values closely track the GAMA SED-derived masses, with a typical scatter of ~0.131 dex. These results demonstrate that complex deep-learning architectures are not a prerequisite for robust stellar mass estimation, and that simulation-trained, lightweight machine-learning models can capture the dominant physical information encoded in broad-band photometry. The method is further applied to 17,006 GAMA galaxies lacking SED-derived masses, with photometric uncertainties propagated through the network to provide corresponding error estimates on the inferred stellar masses. Overall, this work establishes a computationally efficient and conceptually transparent pathway for simulation-to-observation transfer learning in galaxy evolution studies.

研究动机与目标

• 在不完全依赖SED拟合的情况下估算恒星质量的挑战，并强调基于SPS的质量的模型依赖性。
• Demonstrate that a lightweight neural network trained on Shark simulations can predict real galaxy masses using broadband photometry.
• Show that the method can be extended to galaxies without SED-based masses and quantify uncertainties.
• Suggest a practical, efficient pathway for simulation-to-observation transfer learning in galaxy evolution studies.

提出的方法

• 使用一个全连接前馈神经网络，具有单隐藏层，之前在Shark仿真光度上进行过训练。
• 在排除超出Shark训练范围的特征后，使用GAMA的24个可用的绝对星等和颜色索引重新训练网络。
• 通过网络传播光度不确定性，以估计预测恒星质量的不确定性。
• 将ANN预测值与GAMA的SED推导质量进行比较，以评估准确性并识别残留偏差。
• 将训练好的ANN应用于17,006个缺少SED质量的GAMA星系，推导质量估计及传播的不确定性。

Figure 1: Distributions of the 24 broadband absolute magnitudes and colour indices (all in the AB system) used as input features for the GAMA galaxies in this study. In each panel, the red dashed vertical lines indicate the minimum and maximum values spanned by the corresponding features in the Shar

实验结果

研究问题

• RQ1 一个在Shark仿真上训练的简单、单隐藏层ANN是否能从宽带光度准确预测GAMA恒星质量？
• RQ2 在约3.5 dex的恒星质量范围内，仿真训练的质量能否很好地再现SED推导的质量？
• RQ3 相对于SED质量，ANN预测的偏差与散射有多大，是否可以纠正？
• RQ4 该模型能否可靠扩展到没有SED推导质量的星系，并给出现实的不确定性估计？

主要发现

• ANN在约3.5 dex的质量范围内，对GAMA基于SED推导的恒星质量的预测散射约为0.131 dex。
• 存在一个小的、约0.1 dex的系统偏移，可能源于Shark基于的质量映射与SPS基的质量映射之间的差异，可通过平滑残差修正去除。
• 修正后，ANN与一对一关系对齐，保留内在散射。
• 对于没有SED质量的17,006个星系，ANN提供的质量估计具有典型的光度扰动不确定性约0.05 dex，在包含模型散射的情况下总体保守不确定性约0.18 dex。
• 该方法展示了使用轻量级模型和宽带光度实现稳健的从仿真到观测的迁移学习，适用于大规模调查。

Figure 2: Top: Raw ANN-predicted stellar masses compared with SED-derived masses for 71 171 GAMA galaxies. The solid line denotes the one-to-one relation, while the red curve shows a polynomial fit to the data. Middle: Corrected ANN predictions after subtracting the fitted residual trend, removing t