Stochastic gradient descent and fast relaxation to thermodynamic equilibrium: A stochastic control approach
Publication:5020209
DOI10.1063/5.0051796zbMath1491.82014arXiv2103.05096OpenAlexW3135106390MaRDI QIDQ5020209
Lara Neureither, Tobias Breiten, Carsten Hartmann, Upanshu Sharma
Publication date: 4 January 2022
Published in: Journal of Mathematical Physics (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/2103.05096
Numerical optimization and variational techniques (65K10) Feedback control (93B52) Stochastic methods (Fokker-Planck, Langevin, etc.) applied to problems in time-dependent statistical mechanics (82C31) Irreversible thermodynamics, including Onsager-Machlup theory (82C35) Computational molecular dynamics in statistical mechanics (82M37) Monte Carlo methods applied to problems in statistical mechanics (82M31)
Related Items (1)
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Markov semigroups with hypocoercive-type generator in infinite dimensions: ergodicity and smoothing
- Optimal non-reversible linear drift for the convergence to equilibrium of a diffusion
- Accelerating diffusions
- A stochastic control approach to reciprocal diffusion processes
- Asymptotics of the spectral gap with applications to the theory of simulated annealing
- Simulated annealing with time-dependent energy function via Sobolev inequalities
- Roth's theorems for matrix equations with symmetry constraints
- Hypocoercivity in metastable settings and kinetic simulated annealing
- Using perturbed underdamped Langevin dynamics to efficiently sample from probability distributions
- Rates of convergence of the Hastings and Metropolis algorithms
- Adaptive sampling of large deviations
- Mean-field Langevin dynamics and energy landscape of neural networks
- Time scales and exponential trend to equilibrium: Gaussian model problems
- Jarzynski's equality, fluctuation theorems, and variance reduction: mathematical analysis and numerical algorithms
- Couplings and quantitative contraction rates for Langevin dynamics
- Smart sampling and incremental function learning for very large high dimensional data
- Distances between transition probabilities of diffusions and applications to nonlinear Fokker-Planck-Kolmogorov equations
- GENERIC formalism of a Vlasov–Fokker–Planck equation and connection to large-deviation principles
- Irreversible Langevin samplers and variance reduction: a large deviations approach
- A consensus-based model for global optimization and its mean-field limit
- Optimal Steering of a Linear Stochastic System to a Final Probability Distribution, Part I
- Optimal Steering of a Linear Stochastic System to a Final Probability Distribution, Part II
- Fast cooling for a system of stochastic oscillators
- Hypocoercivity
- A Variational Approach to Nonlinear Estimation
- Quantification of coarse-graining error in Langevin and overdamped Langevin dynamics
- Riemann Manifold Langevin and Hamiltonian Monte Carlo Methods
- Rational Construction of Stochastic Numerical Methods for Molecular Sampling
- The Connections Between Lyapunov Functions for Some Optimization Algorithms and Differential Equations
- Stochastic modified equations for the asynchronous stochastic gradient descent
- The Smoothed Spectral Abscissa for Robust Stability Optimization
- Sampling can be faster than optimization
- Data assimilation: The Schrödinger perspective
- The Convergence of Markov Chain Monte Carlo Methods: From the Metropolis Method to Hamiltonian Monte Carlo
This page was built for publication: Stochastic gradient descent and fast relaxation to thermodynamic equilibrium: A stochastic control approach
