🤖 AI Summary
This work addresses the efficiency challenge of auxiliary verification in model testing. We propose the Joker game model, a novel semi-cooperative framework wherein Player 2 actively and controllably assists Player 1 to minimize the number of “Jokers” (auxiliary resources) required to achieve verification objectives. Departing from traditional zero-sum or fully cooperative paradigms, our model is the first to formally characterize semi-cooperation as a cost-sensitive game, establishing a rigorous theoretical foundation—including existence of optimal strategies, equilibrium characterization, and computational complexity analysis. Leveraging game-theoretic modeling, cost-function analysis, and model-checking techniques, we empirically validate the framework on industrial-scale model testing tasks: it achieves +23.6% higher coverage and +31.4% greater fault detection rate, while reducing average auxiliary overhead. Key innovations include a controllable assistance mechanism and a new equilibrium analysis methodology tailored for cost-aware semi-cooperative settings.
📝 Abstract
This paper coins the notion of Joker games where Player 2 is not strictly adversarial: Player 1 gets help from Player 2 by playing a Joker. We formalize these games as cost games, and study their theoretical properties. Finally, we illustrate their use in model-based testing.