A Cosmic Challenge October 30, 2008Posted by Sarah in science.
Tags: astronomy, cosmology, dark energy, gravitational lensing, ucl
Cosmology, the study of the Universe on the very largest of scales, is a frustrating business. The vast majority of the matter in the Universe is unaccounted for, and of a large fraction, which we call Dark Energy, we have no idea what it even might look like, let alone how to find it. One important source of information comes from the study of gravitational lenses. When light from the most distant sources travels across the Universe, it is distorted by intervening matter, as predicted by Einstein’s general theory of relativity. By studying the distortions seen in these distant objects, cosmologists gather information about the properties of the large-scale matter distribution along the line of sight.
But the study of these gravitational lenses is hampered by many problems, from the sheer volumes of data involved, to the instrumental contributions to the shape distortions in these faintest of sources. A large team of gravitational lens cosmologists have now thrown down the gauntlet to computer scientists, statisticians and mathematicians to develop novel algorithms to help with the processing of their highly complex data.
With the GRavitational lEnsing Accuracy Testing 2008 challenge, or GREAT08, a team of researcher s from 19 international institutes led by Dr. Sarah Bridle from University College London are inviting the wider research community to develop new ways of examining gravitational lensing data and extracting statistics of the observed shear in a fast and efficient manner. Not only is this an ‘invitation’, the team have actually set a competition with rules, test data and a leader board! You can watch Sarah Bridle’s screencast to get more background info. A substantive paper was published on astro-ph and in the Annals of Applied Statistics.
This is a really nice idea from the lensing community and I hope it gets some attention. Reserachers in all sciences essentially face many similar problems, particularly when it comes to modelling or number crunching – yet we don’t see much cross-over between the disciplines. UCL is incidentally where I did my PhD and it’s great to see the department taking the lead in this initiative!
So if you’re an expert in statistics, mathematics or computer learning, check out the website with all the competition rules, instructions and links to the data.