Mattias Villani: Optimal Tuning of Subsampling Hamiltonian Monte Carlo,
Tid: On 2018-12-19 kl 15.15 - 16.15
Föreläsare: Mattias Villani (SU),
Plats: Room 306, House 6, Kräftriket, Department of Mathematics, Stockholm University ￼
Abstract: Hamiltonian Monte Carlo (HMC) is an increasingly popular simulation algorithm for Bayesian inference which has proven to be especially suitable in high-dimensional problems. A drawback of HMC is that it requires a large number of evaluations of the posterior gradient, which can be computationally costly, particularly in problems with large datasets. I will present our work on accelerating HMC by data subsampling and our recent results on how to optimally tune the algorithm.