A Unifying and Canonical Description of Measure-Preserving Diffusions
arXiv:2105.02845
Abstract
A complete recipe of measure-preserving diffusions in Euclidean space was recently derived unifying several MCMC algorithms into a single framework. In this paper, we develop a geometric theory that improves and generalises this construction to any manifold. We thereby demonstrate that the completeness result is a direct consequence of the topology of the underlying manifold and the geometry induced by the target measure ; there is no need to introduce other structures such as a Riemannian metric, local coordinates, or a reference measure. Instead, our framework relies on the intrinsic geometry of and in particular its canonical derivative, the deRham rotationnel, which allows us to parametrise the Fokker--Planck currents of measure-preserving diffusions using potentials. The geometric formalism can easily incorporate constraints and symmetries, and deliver new important insights, for example, a new complete recipe of Langevin-like diffusions that are suited to the construction of samplers. We also analyse the reversibility and dissipative properties of the diffusions, the associated deterministic flow on the space of measures, and the geometry of Langevin processes. Our article connects ideas from various literature and frames the theory of measure-preserving diffusions in its appropriate mathematical context.
References in corpus (37)
- A Conceptual Introduction to Hamiltonian Monte Carlo
- Stochastic Gradient Hamiltonian Monte Carlo
- An elementary introduction to information geometry
- Stochastic Climate Theory and Modelling
- Testing and tuning symplectic integrators for Hybrid Monte Carlo algorithm in lattice QCD
- Non-convex learning via Stochastic Gradient Langevin Dynamics: a nonasymptotic analysis
- The computation of averages from equilibrium and nonequilibrium Langevin molecular dynamics
- Optimal non-reversible linear drift for the convergence to equilibrium of a diffusion
- Accelerating diffusions
- Measuring Sample Quality with Stein's Method
- On the Geometric Ergodicity of Hamiltonian Monte Carlo
- Underdamped Langevin MCMC: A non-asymptotic analysis
- Langevin diffusions and the Metropolis-adjusted Langevin algorithm
- Geometric integrators and the Hamiltonian Monte Carlo method
- Stochastic Dynamical Structure (SDS) of Nonequilibrium Processes in the Absence of Detailed Balance. III: potential function in local stochastic dynamics and in steady state of Boltzmann-Gibbs type distribution function
- Proposals which speed-up function-space MCMC
- Compressible Generalized Hybrid Monte Carlo
- Relation of a New Interpretation of Stochastic Differential Equations to Ito Process
- On the convergence of Hamiltonian Monte Carlo
- Modified Hamiltonian Monte Carlo for Bayesian inference
- Is There an Analog of Nesterov Acceleration for MCMC?
- Hamiltonian Monte Carlo Without Detailed Balance
- Shadow Hamiltonians, Poisson Brackets, and Gauge Theories
- Global Convergence of Stochastic Gradient Hamiltonian Monte Carlo for Non-Convex Stochastic Optimization: Non-Asymptotic Performance Bounds and Momentum-Based Acceleration
- Momentum Maps and Stochastic Clebsch Action Principles
- Improving dynamical lattice QCD simulations through integrator tuning using Poisson brackets and a force-gradient integrator
- Nonreversible Langevin Samplers: Splitting Schemes, Analysis and Implementation
- A diffusion approach to Stein's method on Riemannian manifolds
- Bayesian Inference on Matrix Manifolds for Linear Dimensionality Reduction
- Notes on Statistically Invariant States in Stochastically Driven Fluid Flows
- Coordinate-free Stochastic Differential Equations as Jets
- Poisson geometry and first integrals of geostrophic equations
- Deep Markov Chain Monte Carlo
- Constraint-Based Regularization of Neural Networks
- Hamiltonian Monte Carlo on Symmetric and Homogeneous Spaces via Symplectic Reduction
- Note on the geodesic Monte Carlo
- Reversible and non-reversible Markov Chain Monte Carlo algorithms for reservoir simulation problems