EvoQRE: Modeling Bounded Rationality in Safety-Critical Traffic Simulation via Evolutionary Quantal Response Equilibrium
Phu-Hoa Pham, Chi-Nguyen Tran, Duy-Minh Dao-Sy, Phu-Quy Nguyen-Lam, Trung-Kiet Huynh
TL;DR
EvoQRE addresses the gap in traffic simulation by replacing perfect rationality with bounded rationality through Quantal Response Equilibrium (QRE) in general-sum Markov games. It couples Logit-QRE with Entropy-Regularized Replicator Dynamics, establishing a two-timescale convergence with rate $\mathcal{O}(\log k / k^{1/3})$, and extends to continuous action spaces using kernel-density and energy-based approaches. The framework is instantiated in EvoQRE, leveraging a pre-trained world model to produce realistic, safe, and controllable traffic scenarios, with strong empirical results on Waymo Open Motion Dataset and nuPlan benchmarks. Calibrating the rationality parameter $\lambda$ from data enables scenario generation that spans a spectrum of driver types, enhancing both validation fidelity and safety testing capabilities. Overall, EvoQRE provides a principled, scalable tool for AV validation that captures human bounded rationality while delivering interpretable controls over scenario realism and risk profiles.
Abstract
Existing traffic simulation frameworks for autonomous vehicles typically rely on imitation learning or game-theoretic approaches that solve for Nash or coarse correlated equilibria, implicitly assuming perfectly rational agents. However, human drivers exhibit bounded rationality, making approximately optimal decisions under cognitive and perceptual constraints. We propose EvoQRE, a principled framework for modeling safety-critical traffic interactions as general-sum Markov games solved via Quantal Response Equilibrium (QRE) and evolutionary game dynamics. EvoQRE integrates a pre-trained generative world model with entropy-regularized replicator dynamics, capturing stochastic human behavior while maintaining equilibrium structure. We provide rigorous theoretical results, proving that the proposed dynamics converge to Logit-QRE under a two-timescale stochastic approximation with an explicit convergence rate of O(log k / k^{1/3}) under weak monotonicity assumptions. We further extend QRE to continuous action spaces using mixture-based and energy-based policy representations. Experiments on the Waymo Open Motion Dataset and nuPlan benchmark demonstrate that EvoQRE achieves state-of-the-art realism, improved safety metrics, and controllable generation of diverse safety-critical scenarios through interpretable rationality parameters.