Marie Skłodowska-Curie Action ITN SUNDIAL


SUNDIAL (SUrvey Network for Deep Imaging Analysis & Learning) is an ambitious interdisciplinary International Training Network of nine research groups in the Netherlands, Belgium, Germany, Finland, France, UK, Spain, and Italy. Our aim is to develop novel algorithms to study the very large databases coming from current-day telescopes in order to better understand galaxy formation and evolution, and to prepare for the huge missions of the next decade.

SUNDIALThe impact of Big Data, i.e. the exploitation of very large datasets, in for instance commerce, security, environmental monitoring, and experimental research is immense. Despite this success, the task of developing efficient algorithms to mine the ever growing datasets is challenging. Astronomers, who are used to generating large volumes of observational data using high-tech observatories, are historically at the forefront of this field

We will train 14 young scientists in the fields of computer science and astronomy, focusing on techniques of automated learning to answer fundamental questions regarding the evolution of galaxies. While these techniques will lead to major advances in astronomy, we will also promote, in collaboration with industrial partners, much more general applications in society, e.g. in medical imaging and remote sensing.

The collaboration is unique since it will develop a platform for deep symbiosis of two radically different strands of approaches: purely data-driven machine learning and specialist approaches based on techniques developed in astronomy. Young scientists trained with such skills are highly demanded both in research and in business.


These are the objectives of SUNDIAL:

  • Develop algorithms to detect every fainter structures and objects in astronomical data
  • Develop new methods for a robust and quantitative comparison of simulated and observed galaxies in order to test current theories for cosmic structure formation
  • Apply these new numerical tools to the astronomical datasets that are being generated by the members of this training network in order to unravel the physical processes that drive galaxy evolution
  • Prepare the nodes of the network, and the students involved, for the scientific exploitation of the upcoming ultra-large datasets generated by surveys such as EUCLID and LSST.


Role of Ghent University

Two young scientists will be trained in Ghent during their PhD. One will perform detailed numerical simulations of the formation and evolution of dwarf galaxies in a cluster environment. This student will work closely together with other PhDs based in the computer science nodes of this network who are developing new techniques to compare simulations with data. The second PhD student will use multi-wavelength data of the Fornax cluster, ranging from X-ray to radio observations. These data are drawn from ongoing and future surveys of the Fornax cluster (the Fornax Deep Survey, the MeerKAT Fornax HI survey). The student will use novel algorithms to detect faint structures and objects in the data.

Both students will produce data and methods that will find future applications beyond the immediate scope of this ITN.



Prof. Sven De Rijcke
Department Fysica & Sterrenkunde/Physics & Astronomy
Phone number: + 32 9 264 47 91