Data science for Physics and Astronomy

Alan Turing Institute

Alan Turing Institute

Adrian Bevan (Queen Mary University of London (GB)), Benjamin Joachimi (UCL), Daniel Maitre, Darren Price (University of Manchester (GB))

Fundamental research in physics and astronomy routinely produces some of the world's largest and most complex datasets, with huge discovery potential. Recent examples include the direct detection of gravitational waves, the discovery of the Higgs boson, and the identification of earth-like exoplanets. The success of many projects depends critically on the ability to extract information rapidly and/or with high precision and accuracy. These disciplines are therefore prime exploiters of modern data science concepts and algorithms, but they also push the boundaries of data science applications. Challenges include inference from large and low signal-to-noise datasets, mining for rare events/objects, decision/classification with extremely high data rates, machine-learning assisted modelling, extremely large simulations, and much more.

To tackle the data deluge, there is an urgent need for knowledge exchange between research fields, as well as with data science, statistics, and computer science practitioners. Great benefits can be expected from pooling ideas and resources and from efficient lines of communication between these fast-moving fields. Moreover, physics and astronomy researchers need to establish funding lines specifically for data science methodology and applications to remain internationally competitive.

This hands-on workshop will bring together physics and astronomy researchers, data science practitioners, as well as industry partners and public sector funders at The Alan-Turing Institute to discuss current challenges and opportunities in data science for physics and astronomy, launch new interdisciplinary collaborations, and work towards a more permanent forum for data science knowledge exchange.