A Convex Programming Approach for Discrete-Time Markov Decision Processes under the Expected Total Reward Criterion

DOI10.1137/19M1255811zbMath1452.90316arXiv1903.08853OpenAlexW2924486842MaRDI QIDQ5130923

Alexandre Genadot, François Dufour

Publication date: 30 October 2020

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

Full work available at URL: https://arxiv.org/abs/1903.08853

zbMATH Keywords

occupation measureMarkov decision processconstraintsconvex programexpected total reward criterion

Mathematics Subject Classification ID

Convex programming (90C25) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Markov and semi-Markov decision processes (90C40)

Related Items

Extreme Occupation Measures in Markov Decision Processes with an Absorbing State, On Reducing a Constrained Gradual-Impulsive Control Problem for a Jump Markov Model to a Model with Gradual Control Only

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• A convex analytic approach to Markov decision processes
• On dynamic programming: Compactness of the space of policies
• The Expected Total Cost Criterion for Markov Decision Processes under Constraints
• Splitting Randomized Stationary Policies in Total-Reward Markov Decision Processes
• Examples in Markov Decision Processes
• Convexity and Optimization in Banach Spaces
• The Expected Total Cost Criterion for Markov Decision Processes under Constraints: A Convex Analytic Approach
• Stationary Policies in Dynamic Programming Models Under Compactness Assumptions
• Constrained Markov Decision Processes with Expected Total Reward Criteria