Variational approach to rare event simulation using least-squares regression
From MaRDI portal
Publication:5227583
DOI10.1063/1.5090271zbMATH Open1421.62009arXiv1901.09195OpenAlexW2913860142WikidataQ91559998 ScholiaQ91559998MaRDI QIDQ5227583FDOQ5227583
Lorenz Richter, Lara Neureither, Omar Kebiri, Carsten Hartmann
Publication date: 6 August 2019
Published in: Chaos: An Interdisciplinary Journal of Nonlinear Science (Search for Journal in Brave)
Abstract: We propose an adaptive importance sampling scheme for the simulation of rare events when the underlying dynamics is given by a diffusion. The scheme is based on a Gibbs variational principle that is used to determine the optimal (i.e. zero-variance) change of measure and exploits the fact that the latter can be rephrased as a stochastic optimal control problem. The control problem can be solved by a stochastic approximation algorithm, using the Feynman-Kac representation of the associated dynamic programming equations, and we discuss numerical aspects for high-dimensional problems along with simple toy examples.
Full work available at URL: https://arxiv.org/abs/1901.09195
Recommendations
- Regenerative rare events simulation via likelihood ratios
- Efficient simulation for dependent rare events with applications to extremes
- Opportunities of robust regression for variance reduction in discrete event simulation
- An iterative stochastic method for simulating large deviations and rare events
- Rare event simulation via importance sampling for linear SPDE's
- New efficient estimators in rare event simulation with heavy tails
- Optimisation of interacting particle systems for rare event estimation
- Rare Event Estimation for Computer Models
- The transform likelihood ratio method for rare event simulation with heavy tails
rare event simulationstochastic optimal control problemleast-squares regressionimportance sampling scheme
Linear regression; mixed models (62J05) Sampling theory, sample surveys (62D05) Optimal stochastic control (93E20)
Cites Work
- Title not available (Why is that?)
- Stochastic differential equations. An introduction with applications.
- A variational representation for certain functionals of Brownian motion
- Backward stochastic differential equations and partial differential equations with quadratic growth.
- Controlled Markov processes and viscosity solutions
- A regression-based Monte Carlo method to solve backward stochastic differential equations
- Least-Squares Monte Carlo for Backward SDEs
- On the uniqueness of solutions to quadratic BSDEs with convex generators and unbounded terminal conditions
- Linear regression MDP scheme for discrete backward stochastic differential equations under general conditions
- Title not available (Why is that?)
- Title not available (Why is that?)
- Rare Event Simulation of Small Noise Diffusions
- Adaptive Multilevel Splitting for Rare Event Analysis
- Sequential Monte Carlo for rare event estimation
- Connections between stochastic control and dynamic games
- Adaptive importance sampling for control and inference
- Efficient computation of optimal actions
- Importance sampling in rare event simulation
- Optimal control as a graphical model inference problem
- Importance Sampling, Large Deviations, and Differential Games
- Handbooks in operations research and management science: Simulation
- Risk-Sensitive Control on an Infinite Time Horizon
- Exit probabilities and optimal stochastic control
- A forward scheme for backward SDEs
- Optimal control of molecular dynamics using Markov state models
- Path integrals and symmetry breaking for optimal control theory
- Asymptotic analysis of nonlinear stochastic risk-sensitive control and differential games
- Subsolutions of an Isaacs Equation and Efficient Schemes for Importance Sampling
- Asymptotic for the principal eigenvalue and eigenfunction of a nearly first-order operator with large potential
- Importance Sampling for Backward SDEs
- Counterexamples in importance sampling for large deviations probabilities
- Stochastic optimal control via forward and backward stochastic differential equations and importance sampling
- Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
- Risk-sensitivity, large deviations and stochastic control
- References
- Adaptive importance sampling with forward-backward stochastic differential equations
- Title not available (Why is that?)
Cited In (11)
- Solving high-dimensional Hamilton-Jacobi-Bellman PDEs using neural networks: perspectives from the theory of controlled diffusions and measures on path space
- Learning-based importance sampling via stochastic optimal control for stochastic reaction networks
- Approximative Policy Iteration for Exit Time Feedback Control Problems Driven by Stochastic Differential Equations using Tensor Train Format
- Existence of relaxed stochastic optimal control for G-SDEs with controlled jumps
- Overcoming the timescale barrier in molecular dynamics: Transfer operators, variational principles and machine learning
- A Koopman framework for rare event simulation in stochastic differential equations
- Collapse of transitional wall turbulence captured using a rare events algorithm
- $\alpha$-Hypergeometric Uncertain Volatility Models and their Connection to 2BSDEs
- Double-loop importance sampling for McKean-Vlasov stochastic differential equation
- Performance analysis conditioned on rare events: an adaptive simulation scheme
- Title not available (Why is that?)
Uses Software
This page was built for publication: Variational approach to rare event simulation using least-squares regression
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5227583)
