Conditional Value-at-Risk for Reachability and Mean Payoff in Markov Decision Processes

DOI10.1145/3209108.3209176zbMATH Open1497.68312arXiv1805.02946OpenAlexW2963819197MaRDI QIDQ5145337FDOQ5145337

Jan Křetínský, Tobias Meggendorfer

Publication date: 20 January 2021

Published in: Proceedings of the 33rd Annual ACM/IEEE Symposium on Logic in Computer Science (Search for Journal in Brave)

Abstract: We present the conditional value-at-risk (CVaR) in the context of Markov chains and Markov decision processes with reachability and mean-payoff objectives. CVaR quantifies risk by means of the expectation of the worst p-quantile. As such it can be used to design risk-averse systems. We consider not only CVaR constraints, but also introduce their conjunction with expectation constraints and quantile constraints (value-at-risk, VaR). We derive lower and upper bounds on the computational complexity of the respective decision problems and characterize the structure of the strategies in terms of memory and randomization.

Full work available at URL: https://arxiv.org/abs/1805.02946

Mathematics Subject Classification ID

Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.) (60J20) Analysis of algorithms and problem complexity (68Q25) Specification and verification (program logics, model checking, etc.) (68Q60) Markov and semi-Markov decision processes (90C40)

Cited In (5)

• Positivity-hardness results on Markov decision processes
• On Optimizing the Conditional Value-at-Risk of a Maximum Cost for Risk-Averse Safety Analysis
• Title not available (Why is that?)
• Entropic risk for turn-based stochastic games
• MDPs as distribution transformers: affine invariant synthesis for safety objectives

This page was built for publication: Conditional Value-at-Risk for Reachability and Mean Payoff in Markov Decision Processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5145337)