Automated Probabilistic Classification of Transients and Variables

TL;DR

This paper presents a prototype system for automated probabilistic classification of astronomical transients and variables, integrating Bayesian and machine learning methods to optimize follow-up observations in large sky surveys.

Contribution

It introduces a novel methodology combining Bayesian and machine learning classifiers with automated feedback for classifying sky transients and variables.

Findings

Development of a prototype classification system.

Integration of automated follow-up feedback.

Potential application to future large sky surveys.

Abstract

There is an increasing number of large, digital, synoptic sky surveys, in which repeated observations are obtained over large areas of the sky in multiple epochs. Likewise, there is a growth in the number of (often automated or robotic) follow-up facilities with varied capabilities in terms of instruments, depth, cadence, wavelengths, etc., most of which are geared toward some specific astrophysical phenomenon. As the number of detected transient events grows, an automated, probabilistic classification of the detected variables and transients becomes increasingly important, so that an optimal use can be made of follow-up facilities, without unnecessary duplication of effort. We describe a methodology now under development for a prototype event classification system; it involves Bayesian and Machine Learning classifiers, automated incorporation of feedback from follow-up observations, and discriminated or directed follow-up requests. This type of methodology may be essential for the massive synoptic sky surveys in the future.