Dr Deaglan Bartlett has been awarded the Early Career Award (A) by the Royal Astronomical Society recognising early achievement in astronomy.
Dr Bartlett focuses on developing statistical and machine learning methods for cosmology and astrophysics. This includes building high-performance machine learning models capable of analysing the latest highly-detailed surveys of the Universe.
A central focus of his research is symbolic regression (SR), a method that automatically discovers mathematical equations directly from data. Unlike deep neural networks, SR produces simple, understandable expressions that reveal the underlying structure of a system, extrapolate reliably, and run orders of magnitude faster. Dr Bartlett develops new SR techniques that combine statistical methods and machine learning to build symbolic models that achieve the accuracy of neural networks whilst being transparent and easy to use across different scientific applications.
Much of Dr Bartlett’s research is carried out within large international collaborations using machine learning to speed up and improve computer simulations of the Universe. These advances will make it possible to fully exploit upcoming galaxy surveys to probe the nature of dark matter, dark energy, and the formation of structure in the Universe.
‘I am very grateful to the RAS for this recognition,’ he comments. ‘I have benefited greatly from the insights and support of my many collaborators in developing new theoretical and computational tools to extract insight from increasingly complex cosmological data. I am fortunate to work at a time when advances in data, computation and methodology are transforming cosmology, and it is a privilege to contribute to this progress.’