Primal-Dual Regression Approach for Markov Decision Processes with General State and Action Spaces

DOI10.1137/22m1526010arXiv2210.00258MaRDI QIDQ6198082

John G. M. Schoenmakers, Denis Belomestny

Publication date: 20 February 2024

Published in: SIAM Journal on Control and Optimization (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2210.00258

zbMATH Keywords

Markov decision processesreinforcement learningdual representationpseudo regressionStein control functionals

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Monte Carlo methods (65C05) Markov and semi-Markov decision processes (90C40)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Quantitative error estimates for a least-squares Monte Carlo algorithm for American option pricing
• Markov decision processes with applications to finance.
• A pure martingale dual for multiple stopping
• Number of paths versus number of basis functions in American option pricing
• Interpolation of Lipschitz functions
• Linear regression MDP scheme for discrete backward stochastic differential equations under general conditions
• Stratified Regression Monte-Carlo Scheme for Semilinear PDEs and BSDEs with Large Scale Parallelization on GPUs
• Information Relaxations and Duality in Stochastic Dynamic Programs
• Regression Methods for Stochastic Control Problems and Their Convergence Analysis
• General Error Estimates for the Longstaff–Schwartz Least-Squares Monte Carlo Algorithm
• Pathwise Stochastic Optimal Control
• TRUE UPPER BOUNDS FOR BERMUDAN PRODUCTS VIA NON‐NESTED MONTE CARLO
• Pricing American Options: A Duality Approach
• The Covering Radius of Randomly Distributed Points on a Manifold
• High-Dimensional Probability
• Monte Carlo valuation of American options
• Dynamic programming for optimal stopping via pseudo-regression
• DUAL REPRESENTATIONS FOR GENERAL MULTIPLE STOPPING PROBLEMS
• Solving the Dual Problems of Dynamic Programs via Regression
• Primal–dual linear Monte Carlo algorithm for multiple stopping—an application to flexible caps
• Semitractability of optimal stopping problems via a weighted stochastic mesh algorithm

This page was built for publication: Primal-Dual Regression Approach for Markov Decision Processes with General State and Action Spaces