Symmetrically processed splitting integrators for enhanced Hamiltonian Monte Carlo sampling

We construct integrators to be used in Hamiltonian (or Hybrid) Monte Carlo sampling. The new integrators are easily implementable and, for a given computational budget, may deliver five times as many accepted proposals as standard leapfrog/Verlet without impairing in any way the quality of the samples. They are based on a suitable modification of the processing technique first introduced by J.C. Butcher. The idea of modified processing may also be useful for other purposes, like the construction of high-order splitting integrators with positive coefficients.

[1]  Robert I. McLachlan,et al.  On the Numerical Integration of Ordinary Differential Equations by Symmetric Composition Methods , 1995, SIAM J. Sci. Comput..

[2]  Fernando Casas,et al.  A Concise Introduction to Geometric Numerical Integration , 2016 .

[3]  Fernando Casas,et al.  On the necessity of negative coefficients for operator splitting schemes of order higher than two , 2005 .

[4]  Masatoshi Imada,et al.  Monte Carlo Calculation of Quantum Systems. II. Higher Order Correction , 1984 .

[5]  Jesús María Sanz-Serna,et al.  Geometric integrators and the Hamiltonian Monte Carlo method , 2017, Acta Numerica.

[6]  G. Rowlands.  A numerical algorithm for Hamiltonian systems , 1991 .

[7]  J. M. Sanz-Serna,et al.  Markov Chain Monte Carlo and Numerical Differential Equations , 2014 .

[8]  J. M. Sanz-Serna,et al.  Numerical Hamiltonian Problems , 1994 .

[9]  Igor P. Omelyan,et al.  Symplectic analytically integrable decomposition algorithms: classification, derivation, and application to molecular dynamics, quantum and celestial mechanics simulations , 2003 .

[10]  Radford M. Neal.  MCMC Using Hamiltonian Dynamics , 2011, 1206.1901.

[11]  M. A. López-Marcos,et al.  Explicit Symplectic Integrators Using Hessian-Vector Products , 1997, SIAM J. Sci. Comput..

[12]  Jesús María Sanz-Serna,et al.  Multi-stage splitting integrators for sampling with modified Hamiltonian Monte Carlo methods , 2018, J. Comput. Phys..

[13]  Adrian Sandu,et al.  A Hybrid Monte Carlo Sampling Filter for Non-Gaussian Data Assimilation , 2015 .

[14]  M. Aleardi,et al.  Hamiltonian Monte Carlo algorithms for target- and interval-oriented amplitude versus angle inversions , 2020 .

[15]  J. Rosenthal,et al.  Scaling limits for the transient phase of local Metropolis–Hastings algorithms , 2005 .

[16]  Mari Paz Calvo,et al.  The Development of Variable-Step Symplectic Integrators, with Application to the Two-Body Problem , 1993, SIAM J. Sci. Comput..

[17]  J. M. Sanz-Serna,et al.  Adaptive Splitting Integrators for Enhancing Sampling Efficiency of Modified Hamiltonian Monte Carlo Methods in Molecular Simulation. , 2017, Langmuir : the ACS journal of surfaces and colloids.

[18]  Siu A. Chin,et al.  Symplectic integrators from composite operator factorizations , 1997 .

[19]  L. Einkemmer.  Structure preserving numerical methods for the Vlasov equation , 2016, 1604.02616.

[20]  Tore Selland Kleppe,et al.  On the application of improved symplectic integrators in Hamiltonian Monte Carlo , 2018, Commun. Stat. Simul. Comput..

[21]  G. Stoltz,et al.  THEORETICAL AND NUMERICAL COMPARISON OF SOME SAMPLING METHODS FOR MOLECULAR DYNAMICS , 2007 .

[22]  Jesús María Sanz-Serna,et al.  Palindromic 3-stage splitting integrators, a roadmap , 2017, J. Comput. Phys..

[23]  Jesús María Sanz-Serna,et al.  Numerical Integrators for the Hybrid Monte Carlo Method , 2014, SIAM J. Sci. Comput..

[24]  S. Duane,et al.  Hybrid Monte Carlo , 1987 .

[25]  David E. Shaw,et al.  Computationally efficient molecular dynamics integrators with improved sampling accuracy , 2012 .

[26]  J. Butcher.  The effective order of Runge-Kutta methods , 1969 .

[27]  M. Girolami,et al.  Riemann manifold Langevin and Hamiltonian Monte Carlo methods , 2011, Journal of the Royal Statistical Society: Series B (Statistical Methodology).

[28]  Jesús María Sanz-Serna,et al.  Adaptive multi-stage integrators for optimal energy conservation in molecular simulations , 2015, J. Comput. Phys..

[29]  Fernando Casas,et al.  Symplectic Integration with Processing: A General Study , 1999, SIAM J. Sci. Comput..