On decompositions of non-reversible processes
arXiv:2210.03666 · doi:10.1088/1742-6596/2514/1/012007
Abstract
Markov chains are studied in a formulation involving forces and fluxes. First, the iso-dissipation force recently introduced in the physics literature is investigated; we show that its non-uniqueness is linked to different notions of duality giving rise to dual forces. We then study Hamiltonians associated to variational formulations of Markov processes, and develop different decompositions for them.
References in corpus (9)
- On the relation between gradient flows and the large-deviation principle, with applications to Markov chains and diffusion
- Irreversible Langevin samplers and variance reduction: a large deviation approach
- Improving the convergence of reversible samplers
- Canonical structure and orthogonality of forces and currents in irreversible Markov chains
- Acceleration of convergence to equilibrium in Markov chains by breaking detailed balance
- Geometric decomposition of entropy production into excess, housekeeping and coupling parts
- Fluctuation symmetry leads to GENERIC equations with non-quadratic dissipation
- Symmetries and Geometrical Properties of Dynamical Fluctuations in Molecular Dynamics
- Variational structures beyond gradient flows: a macroscopic fluctuation-theory perspective