A Theoretical Framework for Fuzzy Topological Modelling of Stochastic Games

Abstract

Stochastic games constitute an important mathematical framework for analysing dynamic strategic interactions in which the present choices of players influence both immediate payoffs and future probabilistic transitions. However, many real-world strategic situations do not possess sharply defined states, perfectly measurable transition structures, or completely crisp strategic environments. This study develops a theoretical framework for fuzzy topological modelling of stochastic games by integrating finite discounted stochastic games with Chang-type fuzzy topological spaces. In the proposed framework, the state space of a stochastic game is equipped with a fuzzy topology in which fuzzy open sets represent graded strategic neighbourhoods such as stability, risk acceptability, payoff adequacy, and transition reliability. A fuzzy transition-continuity condition is introduced, and a membership-weighted payoff function is defined. The study further formulates a fuzzy Bellman operator and establishes its contraction property in finite discounted zero-sum stochastic games. The study also discusses the existence of stationary fuzzy-topological equilibrium in finite discounted non-zero-sum games. An illustrative three-state stochastic game is used to demonstrate the role of fuzzy openness, ?-cuts, and fuzzy equilibrium stability. The Results section includes data tables and plot specifications to show how fuzzy topological information modifies classical value functions. The framework provides a mathematically tractable approach for studying stochastic games under vagueness, partial admissibility, and graded strategic uncertainty.

Introduction

The text develops a theoretical framework combining fuzzy set theory, fuzzy topology, and stochastic games to model strategic systems in which both randomness and vagueness are present.

The motivation is that classical stochastic games generally assume crisp states, precise transition probabilities, and clearly defined payoffs. While probability captures randomness, it does not adequately represent situations where a state is only partially stable, secure, profitable, cooperative, or acceptable. Fuzzy set theory addresses this by assigning states degrees of membership between 0 and 1.

The proposed fuzzy topological stochastic game equips the state space of a finite discounted stochastic game with a fuzzy topology. Membership functions can represent properties such as:

• payoff adequacy,
• transition reliability,
• risk acceptability,
• strategic stability,
• cooperation or vulnerability.

A fuzzy topology generated from these membership functions creates graded strategic neighbourhoods. The model therefore retains stochastic transitions while allowing states to possess different degrees of strategic suitability.

Main mathematical components

The framework introduces several extensions of classical stochastic-game concepts:

1. Fuzzy strategic neighbourhoods
   States are assigned membership degrees indicating how strongly they satisfy desirable strategic conditions.
2. Fuzzy transition-continuity
   The transition mechanism is required to preserve the structural meaning of fuzzy strategic neighbourhoods as the game evolves probabilistically.
3. Membership-weighted/fuzzy-adjusted payoff
   Conventional payoffs are modified according to the strategic membership of the state, allowing unstable, risky, or weakly admissible states to receive reduced effective payoffs.
4. Fuzzy Bellman operator
   The classical Bellman formulation for finite discounted zero-sum stochastic games is extended by replacing the original payoff with the fuzzy-adjusted payoff.
5. Fuzzy-topological equilibrium analysis
   The framework aims to study equilibrium stability using both stochastic-game dynamics and graded topological properties.

Central theoretical claim

The most important claim is that introducing fuzzy topology does not destroy the mathematical structure underlying finite discounted stochastic games.

In particular, the study argues that the fuzzy Bellman operator remains a contraction for finite discounted zero-sum games. This means that the standard fixed-point and dynamic-programming machinery can still be applied to the proposed fuzzy extension.

The text also begins proving that the family of strategic neighbourhoods generated through arbitrary pointwise suprema and finite pointwise infima satisfies the axioms of a fuzzy topology.

Overall contribution

The paper's contribution is primarily mathematical and theoretical rather than computational. It attempts to provide a unified model in which:

probability → represents randomness in state transitions

fuzzy membership → represents graded strategic uncertainty or suitability

fuzzy topology → represents structured strategic neighbourhoods

dynamic programming → determines long-term game values

The resulting framework could potentially be applied to domains such as cybersecurity, economics, political conflict, risk-sensitive decision-making, and other dynamic strategic systems where states are simultaneously probabilistic and only partially characterized.

Conclusion

This study developed a theoretical framework for fuzzy topological modelling of stochastic games. The model begins with a finite discounted stochastic game and equips its state space with a fuzzy topology generated by membership functions such as payoff adequacy, transition reliability, and risk acceptability. The resulting fuzzy topological stochastic game allows each state to possess a graded level of strategic openness. The study introduced fuzzy transition-continuity, membership-weighted payoff, fuzzy Bellman operator, ?-cut strategic neighbourhoods, and fuzzy-topological equilibrium stability. The main theoretical result established that the fuzzy Bellman operator remains a contraction in finite discounted zero-sum stochastic games. This ensures the existence of a unique fuzzy value function. The study also showed that stationary equilibrium exists in finite discounted non-zero-sum games under bounded fuzzy-adjusted payoffs. The results demonstrated that fuzzy topology changes the interpretation of stochastic game values in a meaningful way. States with weak strategic openness receive lower fuzzy-topological values even when they remain probabilistically relevant. ?-cut analysis further identifies stable strategic regions, while the equilibrium stability index measures graded strategic robustness. The proposed framework contributes to mathematical game theory by combining stochastic dynamics with fuzzy topological structure. It extends the classical state-space approach and provides a foundation for future work on infinite state spaces, measurable fuzzy topologies, fuzzy stochastic games with incomplete information, and computational algorithms for fuzzy equilibrium approximation.

References

[1] L. A. Zadeh, “Fuzzy sets,” Information and Control, vol. 8, no. 3, pp. 338–353, 1965, doi: 10.1016/S0019-9958(65)90241-X. [2] C. L. Chang, “Fuzzy topological spaces,” Journal of Mathematical Analysis and Applications, vol. 24, no. 1, pp. 182–190, 1968, doi: 10.1016/0022-247X(68)90057-7. [3] L. S. Shapley, “Stochastic games,” Proceedings of the National Academy of Sciences of the United States of America, vol. 39, no. 10, pp. 1095–1100, 1953, doi: 10.1073/pnas.39.10.1095. [4] A. M. Fink, “Equilibrium in a stochastic n-person game,” Journal of Science of the Hiroshima University, Series A-I: Mathematics, vol. 28, no. 1, pp. 89–93, 1964. [5] J.-F. Mertens and A. Neyman, “Stochastic games have a value,” Proceedings of the National Academy of Sciences of the United States of America, vol. 79, no. 6, pp. 2145–2146, 1982, doi: 10.1073/pnas.79.6.2145. [6] J. Filar and K. Vrieze, Competitive Markov Decision Processes. New York, NY, USA: Springer, 1997. [7] M. L. Puterman, Markov Decision Processes: Discrete Stochastic Dynamic Programming. Hoboken, NJ, USA: Wiley, 1994. [8] D. Dubois and H. Prade, Fuzzy Sets and Systems: Theory and Applications. New York, NY, USA: Academic Press, 1980. [9] J. F. Nash, “Equilibrium points in n-person games,” Proceedings of the National Academy of Sciences of the United States of America, vol. 36, no. 1, pp. 48–49, 1950, doi: 10.1073/pnas.36.1.48. [10] S. Kakutani, “A generalization of Brouwer’s fixed point theorem,” Duke Mathematical Journal, vol. 8, no. 3, pp. 457–459, 1941, doi: 10.1215/S0012-7094-41-00838-4. [11] J. von Neumann and O. Morgenstern, Theory of Games and Economic Behavior. Princeton, NJ, USA: Princeton University Press, 1944. [12] D. Blackwell, “Discounted dynamic programming,” Annals of Mathematical Statistics, vol. 36, no. 1, pp. 226–235, 1965, doi: 10.1214/aoms/1177700285. [13] L. Attia and M. Oliu-Barton, “A formula for the value of a stochastic game,” Proceedings of the National Academy of Sciences of the United States of America, vol. 116, no. 52, pp. 26435–26443, 2019, doi: 10.1073/pnas.1908643116. [14] E. Solan, “Stochastic games,” Proceedings of the National Academy of Sciences of the United States of America, vol. 112, no. 45, pp. 13743–13746, 2015, doi: 10.1073/pnas.1513508112.

Copyright

Copyright © 2026 Ajai Kumar. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.