Estimating the speed-up of adaptively restrained Langevin dynamics

Trstanova, Zofia; Redon, Stéphane

doi:10.1016/j.jcp.2017.02.010

Cited by 5 publications

(11 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This choice of parameters corresponds to ∼ 30% percent of particles which are frozen for both AR-kinetic energies, i.e. which are in the region where ∇U vanishes (see [42] for a thorough discussion on the link between the percentage of frozen particles and the algorithmic speed-up). Note that the predicted scalings of the rejection rates are recovered in all cases.…”

Section: Average Rejection Ratesmentioning

confidence: 99%

See 1 more Smart Citation

Langevin Dynamics With General Kinetic Energies

Stoltz¹,

Trstanova²

2018

Multiscale Model. Simul.

Self Cite

View full text Add to dashboard Cite

We study Langevin dynamics with a kinetic energy different from the standard, quadratic one in order to accelerate the sampling of Boltzmann-Gibbs distributions. In particular, this kinetic energy can be non-globally Lipschitz, which raises issues for the stability of discretizations of the associated Langevin dynamics. We first prove the exponential convergence of the law of the continuous process to the Boltzmann-Gibbs measure by a hypocoercive approach, and characterize the asymptotic variance of empirical averages over trajectories. We next develop numerical schemes which are stable and of weak order two, by considering splitting strategies where the discretizations of the fluctuation/dissipation are corrected by a Metropolis procedure. We use the newly developped schemes for two applications: optimizing the shape of the kinetic energy for the so-called adaptively restrained Langevin dynamics (which considers perturbations of standard quadratic kinetic energies vanishing around the origin); and reducing the metastability of some toy models using non-globally Lipschitz kinetic energies. arXiv:1609.02891v5 [cond-mat.stat-mech]

show abstract

Section: Average Rejection Ratesmentioning

confidence: 99%

“…The computational gain follows from the fact that the interactions between frozen particles need not be updated. A mathematical analysis of the asymptotic variance for this method is presented in [34], while the algorithmic speed-up, which allows to decrease the cost of a single iteration, is made precise in [42];…”

Section: Introductionmentioning

confidence: 99%

Langevin Dynamics With General Kinetic Energies

Stoltz¹,

Trstanova²

2018

Multiscale Model. Simul.

Self Cite

View full text Add to dashboard Cite

show abstract

“…We note that, as the ARMD methodology has been validated beforehand, the aim of these benchmarks is to validate the single‐pass incremental force update algorithms.…”

Section: Resultsmentioning

confidence: 99%

“…When two particles are restrained, the forces they apply on each other do not need to be updated from one time to the next. As a result, incremental force update algorithms, that is, force calculations that keep track of the total force applied on each particle and update forces involving active particles, may significantly speedup the simulation …”

Section: Introductionmentioning

confidence: 99%

Single‐pass incremental force updates for adaptively restrained molecular dynamics

Singh

Redon

2017

J Comput Chem

Self Cite

View full text Add to dashboard Cite

Adaptively restrained molecular dynamics (ARMD) allows users to perform more integration steps in wall-clock time by switching on and off positional degrees of freedoms. This article presents new, single-pass incremental force updates algorithms to efficiently simulate a system using ARMD. We assessed different algorithms for speedup measurements and implemented them in the LAMMPS MD package. We validated the single-pass incremental force update algorithm on four different benchmarks using diverse pair potentials. The proposed algorithm allows us to perform simulation of a system faster than traditional MD in both NVE and NVT ensembles. Moreover, ARMD using the new single-pass algorithm speeds up the convergence of observables in wall-clock time. © 2017 Wiley Periodicals, Inc.

show abstract

“…Despite the conceptual simplicity of the ARMD approach, however, only simple implementations have been demonstrated so far [18,25], and integrating it in an existing molecular dynamics package (e.g LAAMPS, GROMACS, etc.) is non-trivial.…”

Section: Introductionmentioning

confidence: 99%

Adaptively restrained molecular dynamics in LAMMPS

Singh

Redon

2017

Modelling Simul. Mater. Sci. Eng.

Self Cite

View full text Add to dashboard Cite

Adaptively Restrained Molecular Dynamics (ARMD) is a recently introduced particles simulation method that switches positional degrees of freedom on and off during simulation in order to speed up calculations. In the NVE ensemble, ARMD allows users to trade between precision and speed while, in the NVT ensemble, it makes it possible to compute statistical averages faster. Despite the conceptual simplicity of the approach, however, integrating it in existing molecular dynamics packages is non-trivial, in particular since implemented potentials should a priori be rewritten to take advantage of frozen particles and achieve a speed-up. In this paper, we present novel algorithms for integrating ARMD in LAMMPS, a popular multipurpose molecular simulation package. In particular, we demonstrate how to enable ARMD in LAMMPS without having to re-implement all available force fields. The proposed algorithms are assessed on four different benchmarks, and show how they allow us to speed up simulations up to one order of magnitude.

show abstract

Estimating the speed-up of adaptively restrained Langevin dynamics

Cited by 5 publications

References 24 publications

Langevin Dynamics With General Kinetic Energies

Langevin Dynamics With General Kinetic Energies

Single‐pass incremental force updates for adaptively restrained molecular dynamics

Adaptively restrained molecular dynamics in LAMMPS

Contact Info

Product

Resources

About