Inferring the physical parameters of galaxy populations from their observed properties is one of the key problems in galaxy evolution, and a necessary ingredient in many cosmological analyses. Typical approaches require an explicit or computable form of the likelihood, and significant computational resources. Simulation Based Inference (SBI) approaches circumvent this requirement, and when combined with machine learning methods, significantly reduce the computational cost of inference. Such approaches are necessary to process the millions to billions of sources from the latest generation of instruments.
Simulation based inference is currently being applied to many astrophysical problems which face similar computing barriers. In this workshop, we aim to bring together both experts in SBI as well as galaxy evolution experts with SBI experience, to discuss problems and solutions faced when applying SBI to problems in galaxy evolution. This will be a small, focused workshop to enable plenty of time for discussion and practical sessions.
This is a four day workshop hosted by Bristol Univerity. Catering will be provided throughout. The capacity for the event is 30 and places will be offered on a first-come first-served basis.
For more information see: https://sbi-galev.github.io/2024/