Weighted -contractivity of Langevin dynamics with singular potentials
arXiv:2104.10574 · doi:10.1088/1361-6544/ac4152
Abstract
Convergence to equilibrium of underdamped Langevin dynamics is studied under general assumptions on the potential allowing for singularities. By modifying the direct approach to convergence in pioneered by F. Hérau and developped by Dolbeault, Mouhot and Schmeiser, we show that the dynamics converges exponentially fast to equilibrium in the topologies and , where denotes the invariant probability measure and is a suitable Lyapunov weight. In both norms, we make precise how the exponential convergence rate depends on the friction parameter in Langevin dynamics, by providing a lower bound scaling as . The results hold for usual polynomial-type potentials as well as potentials with singularities such as those arising from pairwise Lennard-Jones interactions between particles.
References in corpus (11)
- Preconditioned Stochastic Gradient Langevin Dynamics for Deep Neural Networks
- Quantitative perturbative study of convergence to equilibrium for collisional kinetic models in the torus
- Hypocoercivity and exponential time decay for the linear inhomogeneous relaxation Boltzmann equation
- Randomized Hamiltonian Monte Carlo
- Spectral Properties of Hypoelliptic Operators
- From ballistic to diffusive behavior in periodic potentials
- Exponential rate of convergence to equilibrium for a model describing fiber lay-down processes
- Non-Equilibrium Steady States for Networks of Oscillators
- A hypocoercivity related ergodicity method with rate of convergence for singularly distorted degenerate Kolmogorov equations and applications
- Geometric ergodicity of Langevin dynamics with Coulomb interactions
- On explicit -convergence rate estimate for piecewise deterministic Markov processes in MCMC algorithms