摘要
We present AGNBoost, a machine learning framework utilizing XGBoostLSS to identify active galactic nuclei (AGNs) and estimate redshifts from James Webb Space Telescope (JWST) Near Infrared Camera (NIRCam) and Mid-Infrared Instrument (MIRI) photometry. AGNBoost constructs 66 input features from seven NIRCam and four MIRI bands to predict the fraction of mid-IR 3-30 mu m emission attributable to an AGN power law (frac(AGN) ) and photometric redshift. Each model is trained on 106 simulated galaxies from the Code Investigating GALaxy Emission (CIGALE). Models are tested on mock CIGALE galaxies, an independent set of empirically derived templates, and 748 observations from the JWST MIRI EGS Galaxy and AGN (MEGA) survey. On idealized noise-free mock CIGALE galaxies, AGNBoost achieves 15% outlier fractions of 1.63% (frac(AGN) ) and 0.15% (redshift), with sigma(RMSE) = 0.045 for frac(AGN) and sigma(NMAD) = 0.004 for redshift. When realistic photometric uncertainties are introduced, performance remains robust with median predictions lying on the 1:1 relation, though outlier fractions increase to 4.38% and 3.35%, respectively. On the independent template set, AGNBoost identifies 92.6% of AGN candidates with frac(AGN) > 0.3 and 100% with frac(AGN) > 0.5, demonstrating generalization beyond the training distribution. For MEGA galaxies with spectroscopic redshifts, AGNBoost achieves sigma(NMAD) = 0.056 and 19.79% outliers. AGNBoost frac(AGN) estimates broadly agree with CIGALE fitting (sigma(RMSE) = 0.178, 11.96% outliers). The flexible framework allows straightforward incorporation of additional photometric bands and retraining for other variables. AGNBoost's computational efficiency makes it well suited for wide-sky surveys requiring rapid AGN identification and redshift estimation.