Deep learning of quasar lightcurves in the LSST era

Deep learning techniques are required for the analysis of synoptic (multi-band and multi-epoch) light curves in massive data of quasars, as expected from the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST). In this follow-up study, we introduced an upgraded version of a conditional neural process (CNP) embedded in a multistep approach for analysis of large data of quasars in the LSST Active Galactic Nuclei Scientific Collaboration data challenge database. We present a case study of a stratified set of the u-band light curves for 283 quasars with very low variability $\sim 0.03$. In this sample, CNP average mean square error is found to be $\sim 5\% $($\sim 0.5$ mag). Interestingly, beside similar level of variability there are indications that individual light curves show flare like features. According to preliminary structure function analysis, these occurrences may be associated to microlensing events with larger time scales $5-10$ years.