Home > Research > Publications & Outputs > Entropic risk for turn-based stochastic games

Research

Links

Text available via DOI:

View graph of relations

Entropic risk for turn-based stochastic games

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Entropic risk for turn-based stochastic games. / Baier, Christel; Chatterjee, Krishnendu; Meggendorfer, Tobias et al.
In: Information and Computation, Vol. 301, 105214, 31.12.2024.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Baier, C, Chatterjee, K, Meggendorfer, T & Piribauer, J 2024, 'Entropic risk for turn-based stochastic games', Information and Computation, vol. 301, 105214. https://doi.org/10.1016/j.ic.2024.105214

APA

Baier, C., Chatterjee, K., Meggendorfer, T., & Piribauer, J. (2024). Entropic risk for turn-based stochastic games. Information and Computation, 301, Article 105214. https://doi.org/10.1016/j.ic.2024.105214

Vancouver

Baier C, Chatterjee K, Meggendorfer T, Piribauer J. Entropic risk for turn-based stochastic games. Information and Computation. 2024 Dec 31;301:105214. Epub 2024 Aug 22. doi: 10.1016/j.ic.2024.105214

Author

Baier, Christel ; Chatterjee, Krishnendu ; Meggendorfer, Tobias et al. / Entropic risk for turn-based stochastic games. In: Information and Computation. 2024 ; Vol. 301.

Bibtex

@article{f21a96b81d7145279e63204989474165,
title = "Entropic risk for turn-based stochastic games",
abstract = "Entropic risk (ERisk) is an established risk measure in finance, quantifying risk by an exponential re-weighting of rewards. We study ERisk for the first time in the context of turn-based stochastic games with the total reward objective. This gives rise to an objective function that demands the control of systems in a risk-averse manner. We show that the resulting games are determined and, in particular, admit optimal memoryless deterministic strategies. This contrasts risk measures that previously have been considered in the special case of Markov decision processes and that require randomization and/or memory. We provide several results on the decidability and the computational complexity of the threshold problem, i.e. whether the optimal value of ERisk exceeds a given threshold. Furthermore, an approximation algorithm for the optimal value of ERisk is provided.",
author = "Christel Baier and Krishnendu Chatterjee and Tobias Meggendorfer and Jakob Piribauer",
year = "2024",
month = dec,
day = "31",
doi = "10.1016/j.ic.2024.105214",
language = "English",
volume = "301",
journal = "Information and Computation",

}

RIS

TY - JOUR

T1 - Entropic risk for turn-based stochastic games

AU - Baier, Christel

AU - Chatterjee, Krishnendu

AU - Meggendorfer, Tobias

AU - Piribauer, Jakob

PY - 2024/12/31

Y1 - 2024/12/31

N2 - Entropic risk (ERisk) is an established risk measure in finance, quantifying risk by an exponential re-weighting of rewards. We study ERisk for the first time in the context of turn-based stochastic games with the total reward objective. This gives rise to an objective function that demands the control of systems in a risk-averse manner. We show that the resulting games are determined and, in particular, admit optimal memoryless deterministic strategies. This contrasts risk measures that previously have been considered in the special case of Markov decision processes and that require randomization and/or memory. We provide several results on the decidability and the computational complexity of the threshold problem, i.e. whether the optimal value of ERisk exceeds a given threshold. Furthermore, an approximation algorithm for the optimal value of ERisk is provided.

AB - Entropic risk (ERisk) is an established risk measure in finance, quantifying risk by an exponential re-weighting of rewards. We study ERisk for the first time in the context of turn-based stochastic games with the total reward objective. This gives rise to an objective function that demands the control of systems in a risk-averse manner. We show that the resulting games are determined and, in particular, admit optimal memoryless deterministic strategies. This contrasts risk measures that previously have been considered in the special case of Markov decision processes and that require randomization and/or memory. We provide several results on the decidability and the computational complexity of the threshold problem, i.e. whether the optimal value of ERisk exceeds a given threshold. Furthermore, an approximation algorithm for the optimal value of ERisk is provided.

U2 - 10.1016/j.ic.2024.105214

DO - 10.1016/j.ic.2024.105214

M3 - Journal article

VL - 301

JO - Information and Computation

JF - Information and Computation

M1 - 105214

ER -