Hybrid Monte Carlo

TL;DR: In this paper, the authors present compelling empirical evidence that the use of penalised regression techniques in the selection of high-dimensional control variates provides performance gains over the classical least squares method.

TL;DR: In this paper, a phenomenological model was proposed to explain the sensitivity of bulk and layered (II,Mn)VI semiconductor optical properties to external magnetic fields, where the low energy degrees of freedom in this model are holes in the semiconductor valence band and one S = 5/2 local moment for each Mn ion.

TL;DR: A new class of metrics aimed at improving HMC's performance on multi-modal target distributions is proposed, referred to as geometrically tempered HMC (GTHMC) due to its connection to other tempering methods and a novel variable step size integrator is developed.

TL;DR: This work introduces Hamiltonian ABC (HABC), a set of likelihood-free algorithms that apply recent advances in scaling Bayesian learning using Hamiltonian Monte Carlo (HMC) and stochastic gradients, and finds that a small number forward simulations can effectively approximate the ABC gradient.

TL;DR: It is argued that the new proposal always represents a significant improvement over a Langevin simulation, and may even improve over the microcanonical method, in which case only a trivial code modification is required.

TL;DR: In this paper, it was shown that the success of acceleration for abelian gauge field dynamics need not depend on any choice of gauge and proposed a particular scheme for acceleration in non-abelian theories which is also gauge independent.

TL;DR: Two algorithms for the numerical simulation of SU(3) lattice gauge theory with dynamical quarks are discussed, based on the hybrid stochastic method of Duane and Kogut, which allow the simulation of arbitrary numbers of quarks.

TL;DR: Fourier techniques that greatly accelerate simulations on large lattices and a new technique for including quark vacuum-polarization corrections that admits any number of flavors, odd or even, without the need for nested Monte Carlo calculations are introduced.