Machine Learning Applications for the Study of AGN Physical Properties Using Photometric Observations

We investigate the physical nature of active galactic nuclei (AGNs) usingmachine learning (ML) tools. We show that the redshift, z, bolometricluminosity, L_ Bol, central mass of the supermassive black hole (SMBH),M_ BH, Eddington ratio, λ_ Edd, and AGN class (obscured orunobscured) can be reconstructed through multi-wavelength photometricobservations only. We trained a random forest regressor (RFR) ML-model on 7616spectroscopically observed AGNs from the SPIDERS-AGN survey, which hadpreviously been cross-matched with soft X-ray observations (from ROSAT or XMM),WISE mid-infrared photometry, and optical photometry from SDSS ugrizfilters. We built a catalog of 21050 AGNs that were subsequently reconstructedwith the trained RFR; for 9687 sources, we found archival redshiftmeasurements. All AGNs were classified as either type 1 or type 2 using arandom forest classifier (RFC) algorithm on a subset of known sources. Allknown photometric measurement uncertainties were incorporated via asimulation-based approach. We present the reconstructed catalog of 21050 AGNswith redshifts ranging from 0 < z < 2.5. We determined z estimations for11363 new sources, with both accuracy and outlier rates within 2distinction between type 1 or type 2 AGNs could be identified with respectiveefficiencies of 94classification, of all sources is given in the dataset. The L_ Bol,M_ BH, and λ_ Edd values are given for 21050 new sourceswith their estimated error. These results have been made publicly available.The release of this catalog will advance AGN studies by presenting keyparameters of the accretion history of 6 dex in luminosity over a wide range ofz.