Prediction of Star Formation Rates Using an Artificial Neural Network
Authors:
Ashraf Ayubinia,
Jong-hak Woo,
Fatemeh Hafezianzadeh,
Taehwan Kim,
Changseok Kim
Abstract:
In this study, we develop an artificial neural network to estimate the infrared (IR) luminosity and star formation rates (SFR) of galaxies. Our network is trained using 'true' IR luminosity values derived from modeling the IR spectral energy distributions (SEDs) of FIR-detected galaxies. We explore five different sets of input features, each incorporating optical, mid-infrared (MIR), near-infrared…
▽ More
In this study, we develop an artificial neural network to estimate the infrared (IR) luminosity and star formation rates (SFR) of galaxies. Our network is trained using 'true' IR luminosity values derived from modeling the IR spectral energy distributions (SEDs) of FIR-detected galaxies. We explore five different sets of input features, each incorporating optical, mid-infrared (MIR), near-infrared (NIR), ultraviolet (UV), and emission line data, along with spectroscopic redshifts and uncertainties. All feature sets yield similar IR luminosity predictions, but including all photometric data leads to slightly improved performance. This suggests that comprehensive photometric information enhances the accuracy of our predictions. Our network is applied to a sample of SDSS galaxies defined as unseen data, and the results are compared with three published catalogs of SFRs. Overall, our network demonstrates excellent performance for star-forming galaxies while we observe discrepancies in composite and AGN samples. These inconsistencies may stem from uncertainties inherent in the compared catalogs or potential limitations in the performance of our network.
△ Less
Submitted 23 December, 2024;
originally announced December 2024.
Exploring galactic properties with machine learning Predicting star formation, stellar mass, and metallicity from photometric data
Authors:
F. Z. Zeraatgari,
F. Hafezianzadeh,
Y. -X. Zhang,
A. Mosallanezhad,
J. -Y. Zhang
Abstract:
Aims. We explore machine learning techniques to forecast star formation rate, stellar mass, and metallicity across galaxies with redshifts ranging from 0.01 to 0.3.
Methods. Leveraging CatBoost and deep learning architectures, we utilize multiband optical and infrared photometric data from SDSS and AllWISE, trained on the SDSS MPA-JHU DR8 catalogue.
Results. Our study demonstrates the potentia…
▽ More
Aims. We explore machine learning techniques to forecast star formation rate, stellar mass, and metallicity across galaxies with redshifts ranging from 0.01 to 0.3.
Methods. Leveraging CatBoost and deep learning architectures, we utilize multiband optical and infrared photometric data from SDSS and AllWISE, trained on the SDSS MPA-JHU DR8 catalogue.
Results. Our study demonstrates the potential of machine learning in accurately predicting galaxy properties solely from photometric data. We achieve minimised root mean square errors, specifically employing the CatBoost model. For star formation rate prediction, we attain a value of RMSESFR = 0.336 dex, while for stellar mass prediction, the error is reduced to RMSESM = 0.206 dex. Additionally, our model yields a metallicity prediction of RMSEmetallicity = 0.097 dex.
Conclusions. These findings underscore the significance of automated methodologies in efficiently estimating critical galaxy properties, amid the exponential growth of multi-wavelength astronomy data. Future research may focus on refining machine learning models and expanding datasets for even more accurate predictions.
△ Less
Submitted 24 May, 2024;
originally announced May 2024.
