On Periodic and Aperiodic Optimal Strategies in Solvency Games
Abstract
Solvency games are a gambling problem on infinite-state Markov decision processes in which the state n ∈ ℕ represents an investor's fortune. In every round, the investor chooses an action from a finite action set, and every action yields a distribution over integer-valued gains in an interval {−ℓ,…,m}. The risk-averse investor wants to minimise the probability of eventual ruin (reaching a fortune ≤ 0). It was shown in [Berger et al.] that memoryless deterministic optimal strategies exist, but they are not eventually constant in general. Even in the special case of gains in {−2,…,1}, the optimal strategy may need to make use of two different actions at arbitrarily high fortunes. We show that optimal strategies in solvency games need not be ultimately periodic in general (thus disproving a 2012 conjecture of Kučera). Already in the case of gains in {−3,…,1}, it is possible for the optimal strategy to be unique but aperiodic. For gains in {−2,…,1}, there always exists an ultimately periodic optimal strategy whose tail is constant or alternates between two actions. Finally, we show that the optimal strategy is computable if it is unique. Moreover, (some) optimal strategy can always be computed in the case of gains in {−ℓ,…,1} for any ℓ ∈ ℕ. Computability in the general case however remains open.
Related
- Infinite Subtraction Games with Periodic Outcomes and Aperiodic SG Values
- Purely Periodic Three-move Subtraction Games
- Optimal Average Distortion for Non-Expansive Embeddings of Euclidean Metrics into the Line: With applications to cut-matching games and vertex expansion
- Receding Horizon Control and Dissipativity - Optimal Control, Games and Uncertainty
- New strategies for some pursuit-evasion differential games with Gronwall-type constraints
BibTeX
@misc{guilmant2026periodic,
title = {{On Periodic and Aperiodic Optimal Strategies in Solvency Games}},
author = {Quentin Guilmant and Florian Luca and Richard Mayr and Joël Ouaknine and James Worrell},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.19438},
eprint = {2609.19438},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.19438},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - On Periodic and Aperiodic Optimal Strategies in Solvency Games
AU - Quentin Guilmant
AU - Florian Luca
AU - Richard Mayr
AU - Joël Ouaknine
AU - James Worrell
PY - 2026
DA - 2026/09/16
PB - arXiv
DO - 10.48550/arxiv.2609.19438
UR - https://arxiv.org/abs/2609.19438
AB - Solvency games are a gambling problem on infinite-state Markov decision processes in which the state n ∈ ℕ represents an investor's fortune. In every round, the investor chooses an action from a finite action set, and every action yields a distribution over integer-valued gains in an interval {−ℓ,…,m}. The risk-averse investor wants to minimise the probability of eventual ruin (reaching a fortune ≤ 0). It was shown in [Berger et al.] that memoryless deterministic optimal strategies exist, but they are not eventually constant in general. Even in the special case of gains in {−2,…,1}, the optimal strategy may need to make use of two different actions at arbitrarily high fortunes. We show that optimal strategies in solvency games need not be ultimately periodic in general (thus disproving a 2012 conjecture of Kučera). Already in the case of gains in {−3,…,1}, it is possible for the optimal strategy to be unique but aperiodic. For gains in {−2,…,1}, there always exists an ultimately periodic optimal strategy whose tail is constant or alternates between two actions. Finally, we show that the optimal strategy is computable if it is unique. Moreover, (some) optimal strategy can always be computed in the case of gains in {−ℓ,…,1} for any ℓ ∈ ℕ. Computability in the general case however remains open.
ER -