Abstract:This paper addresses complexity problems in rational verification and synthesis for multi-player games played on weighted graphs, where the objective of each player is to minimize the cost of reaching a specific set of target vertices. In these games, one player, referred to as the system, declares his strategy upfront. The other players, composing the environment, then rationally make their moves according to their objectives. The rational behavior of these responding players is captured through two models: they opt for strategies that either represent a Nash equilibrium or lead to a play with a Pareto-optimal cost tuple.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the complexity problem of rational verification and synthesis in multi - player games, especially in games played on weighted graphs. Each player's goal is to minimize the cost of reaching a specific set of target vertices. In this context, the system player (referred to as "system") first announces its strategy, and other players in the environment react rationally according to their goals. These rational behaviors are captured by two models: one is the strategy representing Nash equilibrium (NE), and the other is the strategy leading to Pareto - optimal (PO) cost tuples. Specifically, the paper studies the following types of problems: 1. **Rational Verification**: - **Non - Cooperative Nash Verification (NCNV)**: Given a threshold \( c \), verify whether the system player's strategy \( \sigma_0 \) can ensure that under any Nash equilibrium formed by the rational responses of environmental players, the system's cost does not exceed \( c \). - **Universal Non - Cooperative Nash Verification (UNCNV)**: For all possible system strategies \( \sigma_0 \), verify whether the above condition holds. - **Non - Cooperative Pareto Verification (NCPV)**: Similarly, but here the rational response refers to the Pareto - optimal game result. - **Universal Non - Cooperative Pareto Verification (UNCPV)**: For all possible system strategies \( \sigma_0 \), verify whether the system's cost does not exceed \( c \) under the Pareto - optimal condition. 2. **Rational Synthesis**: - **Cooperative Nash Synthesis (CNS)**: Synthesize a system strategy \( \sigma_0 \) such that there exists a Nash equilibrium \( \sigma \) in which the system's cost does not exceed \( c \). - **Non - Cooperative Nash Synthesis (NCNS)**: Synthesize a system strategy \( \sigma_0 \) such that for all Nash equilibria \( \sigma \), the system's cost does not exceed \( c \). - **Cooperative Pareto Synthesis (CPS)**: Similarly, but here the rational response refers to the Pareto - optimal game result. - **Non - Cooperative Pareto Synthesis (NCPS)**: Synthesize a system strategy \( \sigma_0 \) such that for all Pareto - optimal game results, the system's cost does not exceed \( c \). ### Main Contributions The main contributions of the paper include: - **Complexity Results**: Determine the exact complexity of most of the research problems, for example: - The non - cooperative Pareto verification problem is \( \Pi_2^P \)-complete. - The universal non - cooperative Pareto verification problem is PSPACE - complete. - The cooperative Pareto synthesis problem is PSPACE - complete. - The non - cooperative Nash verification problem is coNP - complete. - The universal non - cooperative Nash verification problem is coNP - complete. - The cooperative Nash synthesis problem is NP - complete. - The non - cooperative Nash synthesis problem is EXPTIME - hard, even in a two - player environment; for a single - player environment, it is in EXPTIME and PSPAC

As Soon as Possible but Rationally

Rational verification and checking for Nash and subgame-perfect equilibria in graph games

The Non-Cooperative Rational Synthesis Problem for Subgame Perfect Equilibria and omega-regular Objectives

Stackelberg-Pareto Synthesis

Stackelberg-Pareto Synthesis with Quantitative Reachability Objectives

Safe Equilibrium

Reactive Synthesis Without Regret

Verification with Common Knowledge of Rationality for Graph Games

Risk-Averse Equilibrium for Games

Rationalizing Irrational Beliefs

Rational Dynamics And Epistemic Logic In Games (Vol 9, Pg 13, 2007)

Multi-weighted Reachability Games and Their Application to Permissiveness

Non-cooperative rational synthesis problem for probabilistic strategies

Rational Dynamics and Epistemic Logic in Games

Reconciling Rationality and Stochasticity: Rich Behavioral Models in Two-Player Games

Approximating Nash Equilibrium in Random Graphical Games

Integrated Resource Allocation and Strategy Synthesis in Safety Games on Graphs with Deception

From First-Order to Second-Order Rationality: Advancing Game Convergence with Dynamic Weighted Fictitious Play

Playing Against Fair Adversaries in Stochastic Games with Total Rewards

On Sparse Discretization for Graphical Games

Probabilistic Model Checking for Strategic Equilibria-based Decision Making: Advances and Challenges