COLLOQUIUM: Paula Sánchez Sáez, ESO Germany

Searching for different AGN populations in massive datasets with Machine Learning

Brightness variations of active galactic nuclei (AGNs) offer key insights into their physical emission mechanisms and related phenomena. These variations also provide us with an alternative way to identify AGN candidates that could be missed by more traditional selection techniques. The 4MOST Chilean AGN/Galaxy Evolution Survey (ChANGES) is taking advantage of this variable behavior to select diverse AGN populations from multiple time domain photometric surveys, including the Zwicky Transient Facility (ZTF), La Silla QUEST survey (LSQ), and the upcoming Rubin Observatory Legacy Survey of Space and Time (LSST). In this talk, I will present the algorithms that ChANGES is using to select low-mass and low-Eddington Rate AGNs, as well as changing-look AGNs (CLAGNs) at different stages of the transition, which will be followed up by 4MOST. I will first present a new variability and color-based classifier, designed to identify multiple classes of transients, persistently variable, and non-variable sources, from these surveys. The main motivation of this model is to identify AGN candidates, but it can be used for more general time-domain astronomy studies. I will show how we are using this model to identify different classes of AGNs and to select CLAGNs that transitioned from type 2 to type 1. Then, I will present a deep learning anomaly detection technique designed to identify AGN light curves with anomalous behaviors in massive datasets, like the ZTF data releases. The main aim of this technique is to identify CLAGNs at different stages of the transition, but it can also be used for more general purposes, such as cleaning massive datasets for AGN variability analyses. 

Talk in presence and via Zoom.

Zoom:

https://uni-hamburg.zoom.us/j/69812366862?pwd=TWhoQ2gxSDlxdm85dmMwcHFSNE1KUT09