BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:GPU Accelerated Nested Sampling - Will Handley (University of Camb
 ridge)
DTSTART:20250627T104500Z
DTEND:20250627T111500Z
UID:TALK232357@talks.cam.ac.uk
DESCRIPTION:Nested Sampling is a Monte Carlo method that performs paramete
 r estimation and model comparison robustly for a variety of high dimension
 al and complicated distributions. It has seen widespread usage in the phys
 ical sciences\, however in recent years increasingly it is viewed as part 
 of a legacy code base\, with GPU native paradigms such as neural simulatio
 n based inference coming to the fore. In this work we demonstrate that we 
 can effectively reformulate Nested Sampling to a form that is highly amena
 ble to modern GPU hardware\, taking unique advantage of vectorization oppo
 rtunities to accelerate numerical inference to state of the art levels. We
  provide a public implementation of this code\, distributed via the blackj
 ax statistical framework\, which allows direct comparison with other well-
 established statistical methods such as Hamiltonian Monte Carlo and Sequen
 tial Monte Carlo\, and in this contribution will explore its application t
 o a number of inference problems such as Gravitational Wave parameter esti
 mation and CMB cosmology.Co-author: David Yallup
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR