COmputational Statistics and MOlecular Simulation

2014 – 2018

This project is supported by the Agence Nationale de la Recherche, under grant ANR-14-CE23-0012


Scientific scope:

This project aims at developing numerical techniques dedicated to the sampling of high-dimensional probability measure describing a system of interest. There are two application fields of interest: computational statistical physics (a field also known as molecular simulation), and computational statistics. These two fields share some common history, but it seems that, in view of the quite recent specialization of the scientists and the techniques used in these respective fields, the communication between molecular simulation and computational statistics is not as intense as it should be.

We believe that there are therefore many opportunities in considering both fields at the same time: in particular, the adaption of a successful simulation technique from one field to the other requires first some abstraction process where the features specific to the original field of application are discarded and only the heart of the method is kept. Such a cross-fertilization is however only possible if the techniques developed in a specific field are sufficiently mature: this is why some fundamental studies specific to one of the application fields are still required. Our belief is that the embedding in a more general framework of specific developments in a given field will accelerate and facilitate the diffusion to the other field.

The main topics are the following:

  • Free energy based adaptive importance sampling;
  • Sampling reactive trajectories;
  • Dynamical behavior of Metropolis-Hastings algorithms;
  • Introducing non-reversible perturbations to enhance the exploration power of equilibrium dynamics.


Some events related to the project and (co)organized by members of it:


  • Colvars (implementation of reaction coordinate functions for free energy computations)

Publications of the project (by topics as listed in the proposal):

Review article

  • T. Lelievre and G. Stoltz, Partial differential equations and stochastic methods in molecular dynamics, Acta Numerica (2016) (HAL)

Adaptive importance sampling: convergence, algorithmic improvements and popularization

  • G. Fort, B. Jourdain, T. Lelievre and G. Stoltz, Self-Healing Umbrella Sampling: Convergence and efficiency, Stat. Comput. (2015) (HAL)
  • G. Fort, B. Jourdain, T. Lelievre and G. Stoltz, Convergence and efficiency of adaptive importance sampling techniques with partial biasing, J. Stat. Phys (2018) (HAL)
  • A. Lesage, T. Lelievre, G. Stoltz and J. Henin, Smoothed biasing forces yield unbiased free energies with the extended-system adaptive biasing force method, accepted in J. Phys. Chem. (2016) (HAL)

Sampling reactive trajectories

  • F. Cerou and A. Guyader, Fluctuation Analysis of Adaptive Multilevel Splitting, Annals of Applied Probability (2016) (HAL)
  • F. Cerou, B. Delyon, A. Guyader and M. Rousset, A central limit theorem for Fleming-Viot particle systems, arXiv preprint 1611.00515 (2016) (HAL)

Dynamical behavior of Metropolis-Hastings algorithms

  • M. Fathi and G. Stoltz, Improving dynamical properties of stabilized discretizations of overdamped Langevin dynamics, Numer. Math. (2016) (HAL)
  • A. Durmus and E. Moulines, Non-asymptotic convergence analysis for the Unadjusted Langevin Algorithm, Ann. Appl. Prob. (2016) (HAL)
  • A. Durmus, S. Le Corff, E. Moulines, G. O. Roberts, Optimal scaling of the Random Walk Metropolis algorithm under Lp mean differentiability, arXiv preprint 1604.06664 (2016) (HAL)
  • T. Lelièvre, M. Rousset and G. Stoltz, Hybrid Monte Carlo methods for sampling probability measures on submanifolds, arXiv preprint 1807.02356 (2018) (HAL)

Variance reduction for non-reversible dynamics and study of nonequilibrium dynamics in general:

  • R. Assaraf, B. Jourdain, T. Lelievre, R. Roux, Computation of sensitivities for the invariant measure of a parameter dependent diffusion, arXiv preprint 1509.01348 (HAL)
  • G. Stoltz and Z. Trstanova, Stable and accurate schemes for Langevin dynamics with general kinetic energies, SIAM MMS (2018) (HAL)
  • S. Redon, G. Stoltz and Z. Trstanova, Error Analysis of Modified Langevin Dynamics, J. Stat. Phys. (2016) (HAL)
  • A. Iacobucci, S. Olla and G. Stoltz, Convergence rates for nonequilibrium Langevin dynamics, Ann. Math. Quebec (2017) (HAL)
  • J. Roussel and G. Stoltz, Spectral methods for Langevin dynamics and associated error estimates, M2AN (2017) (HAL)
  • G. Stoltz and E. Vanden-Eijnden, Longtime convergence of the Temperature-Accelerated Molecular Dynamics Method, Nonlinearity (2018) (HAL)
  • G. Ferré and G. Stoltz, Error estimates on ergodic properties of discretized Feynman-Kac semigroups, arXiv preprint 1712.04013 (2016) (HAL)
  • J. Roussel and G. Stoltz, A perturbative approach to control variates in molecular dynamics, arXiv preprint 1712.08022 (2016) (HAL)

Model reduction and effective dynamics:

  • G. Faure and G. Stoltz, Stable and accurate schemes for smoothed dissipative particle dynamics, Appl. Math. Mech.-Engl. (2018) (HAL)
  • G. Stoltz, Stable schemes for dissipative particle dynamics with conserved energy, J. Comput. Phys. (2017) (HAL)
  • G. Faure, J. Roussel, J.-B. Maillet, G. Stoltz, Size consistency in Smoothed Dissipative Particle Dynamics, Phys. Rev. E. (2016) (HAL)
  • A.-A. Homman, J.-B. Maillet, J. Roussel and G. Stoltz, New parallelizable schemes for integrating the Dissipative Particle Dynamics with Energy Conservation, J. Chem. Phys (2016) (HAL)