Human game experiment to verify the equilibrium selection controlled by design
Wang Zhijian, Shan Lixia, Yao Qinmei, Wang Yijia

TL;DR
This study experimentally verifies that equilibrium selection in games can be influenced by game dynamics design, with human behavior aligning with evolutionary game theory predictions, demonstrating the potential for controlled equilibrium outcomes.
Contribution
It provides empirical evidence that manipulating game dynamics can steer equilibrium selection, bridging control theory and human strategic behavior.
Findings
Human strategies follow evolutionary game theory predictions
Cyclic patterns observed in strategy space
Convergence speed aligns with theoretical expectations
Abstract
We conducted a laboratory experiment involving human subjects to test the theoretical hypothesis that equilibrium selection can be impacted by manipulating the games dynamics process, by using modern control theory. Our findings indicate that human behavior consists with the predictions derived from evolutionary game theory paradigm. The consistency is supported by three key observations: (1) the long-term distribution of strategies in the strategy space, (2) the cyclic patterns observed within this space, and (3) the speed of convergence to the selected equilibrium. These findings suggest that the design of controllers aimed at equilibrium selection can indeed achieve their theoretical intended purpose. The location of this study in the knowledge tree of evolutionary game science is presented.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsReinforcement Learning in Robotics · Evolutionary Algorithms and Applications · Artificial Intelligence in Games
MethodsSPEED: Separable Pyramidal Pooling EncodEr-Decoder for Real-Time Monocular Depth Estimation on Low-Resource Settings
