Astronomical observations already produce vast amounts of data through a new generation of telescopes (Atacama Large Millimeter Array (ALMA), Jansky VLA) and through large surveys (e.g., SDSS [1], ZTF [2], PanSTARRS [3], VLT Survey Telescope -VST, and many others) that cannot be analyzed manually. Next-generation telescopes such as the Large Synoptic Survey Telescope (LSST [4]) and the Square Kilometer Array (SKA [5]) are planned to become operational in this decade and the next, and will increase the data volume by many orders of magnitude. The increased spatial, temporal and spectral resolution afford a powerful magnifying lens on the physical processes that underlie the data but, at the same time, generate unprecedented complexity hard to exploit for knowledge extraction. It is therefore imperative to develop machine intelligence, machine learning (ML) in particular, suitable for processing the amount and variety of astronomical data that will be collected, and capable of answering scientific questions based on the data [6].Astronomical data exhibit the usual challenges associated with big data such as immense volumes, high dimensionality, missing or highly distorted observations. In addition, astronomical data can exhibit large continuous observational gaps, very low signal-to-noise ratio and the need to distinguish between true missing (i.e., noncollected) data and non-detections (i.e., due to upper limits). There are strict laws of physics behind the data production which can be assimilated into ML mechanisms to improve over general off-the-shelf state-of-the-art methods. An additional peculiarity is that these large and heterogeneous data sets [7] need to be simultaneously queued, merged and mined by many independent groups of researchers, posing problems not common in many other application domains. In this context it is important to mention