SUNDIAL - SUrvey Network for Deep Imaging Analysis and Learning
SUNDIAL (SUrvey Network for Deep Imaging Analysis & Learning) is an ambitious interdisciplinary network of nine research groups in the Netherlands, Germany, Finland, France, the United Kingdom, Spain, Belgium and Italy. The aim of the network is to develop novel algorithms to study the very large databases coming from current-day telescopes to better understand galaxy formation and evolution, and to prepare for the huge missions of the next decade.
Big Data has become common nowadays but it is a challenge to develop efficient and automated use of the ever-increasing data sets by new generations of data scientists. The project aims to contribute to this general discussion by training a number of young scientists in the fields of computer science and astronomy, focussing on techniques of automated learning from large quantities of data to answer fundamental questions on the evolution of properties of galaxies.
It will also promote, in collaboration with industry, much more general application of big data in society, e.g. in medical imaging or remote sensing.
Basic information
- The University of Birmingham, UoB, United Kingdom
- Ruprecht-Karlsuniversitaet, UHEI Heidelberg, Germany
- Oulun Yliopisto, Uoulu, Finland
- Chambre de Commerce et d'industrie de region Paris Ile-De-France, ESIEE France
- Universita degli Studi di Napoli Federico II, Italy
- Istituto Nazionale Di Astrofisica, INAF, Italy
- Instituto de Astrofisica de Canarias IAC, Spain
- Universiteit Gent UGent, Belgium
The project has put together a team of astronomers and computer scientists, from academic and private sector partners, to develop techniques to detect and classify ultra-faint galaxies and galaxy remnants in a deep survey of the Fornax cluster, and use the results to study how galaxies evolve in the dense environment of galaxy clusters.
With a team of young researchers SUNDIAL will develop novel computer science algorithms addressing fundamental topics in galaxy formation, such as the huge dark matter fractions inferred by theory, and the lack of detected angular momentum in galaxies. The collaboration is unique – 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 in both research and business.
The research in the project is divided into three work packages:
- Automated Detection Methods of Faint Structures
- Quantitative Galaxy Classification and Automated Exploratory Analysis
- Relating simulations with observations- characterisation and visualisation of simulations in the context of observations and vice versa
The SUNDIAL project has published a set of online teaching materials and accompanying tools which were developed in collaboration with beneficiary IAC under its 30th Canary Islands Winter School programme.
In its 2018 edition, this annual school was centred on the theme of Big Data Analysis in Astronomy, and was aimed at early-stage researchers in computer sciences and astronomy. The school was co-organised by the SUNDIAL ITN and a significant number of the SUNDIAL ESRs participated. In total, over 80 young researchers from all continents attended the week-long training course, during which a world-leading team of five instructors delivered a set of lectures, tutorials and practical sessions. The topics and lecturers were:
- General overview on the use of machine learning techniques in astronomy, Prof. S. George Djorgovski (Caltech, Division of Physics, Mathematics and Astronomy),
- Data challenges and solutions in forthcoming surveys, Prof. Mario Juric (University of Washington),
- Machine learning methods for non-supervised classification and dimension reduction techniques, Dalya Baron (School of Physics and Astronomy, Tel-Aviv University),
- Supervised learning: classification and regression, Prof. Michael Biehl (Johann Bernoulli Institute for Mathematics and Computer Science, University of Groningen),
- Deep learning, Prof. Marc Huertas-Company (Université Paris-Diderot - Observatoire de Paris and Instituto de Astrofísica de Canarias).
The material has been published on the following websites:
For the Sundial project Manou Ijtsma has developed a lesson series of five lessons about astronomical topics for students aged 8-12. The lessons can be given by any teacher, even if you have no knowledge about astronomy. The lessons can be downloaded ready to use with PowerPoint slides to illustrate several different topics, such as Mars, the Milky Way galaxy and exoplanets.
The information which can be taught during these lessons is available in the notes and there are even some extra materials in these lessons which could be added for advanced groups. These lessons are an ideal way for teachers to teach about astronomy without hours of preparing lessons. Astronomy is one of those topics which get the attention of students almost immediately and the easiness of giving these lessons makes you want to give them again every year.