The Acceptance Probability of the Hybrid Monte Carlo Method in High-Dimensional Problems.
In: Simos, TE and Psihoyios, G and Tsitouras, C, (eds.)
NUMERICAL ANALYSIS AND APPLIED MATHEMATICS, VOLS I-III.
(pp. 23 - 26).
AMER INST PHYSICS
We investigate the properties of the Hybrid Monte-Carlo algorithm in high dimensions. In the simplified scenario of independent, identically distributed components, we prove that, to obtain an 0(1) acceptance probability as the dimension d of the state space tends to infinity, the Verlet/leap-frog step-size h should be scaled as h = l x d(-1/4). We also identify analytically the asymptotically optimal acceptance probability, which turns out to be 0.651 (with three decimal places); this is the choice that optimally balances the cost of generating a proposal, which decreases as l increases, against the cost related to the average number of proposals required to obtain acceptance, which increases as increases.
|Title:||The Acceptance Probability of the Hybrid Monte Carlo Method in High-Dimensional Problems|
|Event:||International Conference on Numerical Analysis and Applied Mathematics|
|Dates:||2010-09-19 - 2010-09-25|
|Keywords:||Hybrid Monte Carlo, Hamiltonian dynamics, Verlet method, Geometric integration|
|UCL classification:||UCL > School of BEAMS > Faculty of Maths and Physical Sciences
UCL > School of BEAMS > Faculty of Maths and Physical Sciences > Statistical Science
Archive Staff Only