Chengyuan Zhang

Postdoctoral Researcher, McGill University — Bayesian learning, trustworthy AI, and driver world models for multi-agent traffic systems.

I am a Postdoctoral Researcher in Civil Engineering at McGill University, working with Prof. Lijun Sun. I received my Ph.D. from the same group in 2026, and my B.Eng. in Vehicle Engineering from Chongqing University in 2019. I was also a visiting researcher at Carnegie Mellon University with Prof. Changliu Liu in 2023 and Prof. Ding Zhao in 2018, and at UC Berkeley with Prof. Masayoshi Tomizuka from 2019 to 2020.

My research builds trustworthy AI for multi-agent traffic systems, combining Bayesian statistics with generative models to create interpretable driver world models that capture the stochasticity of human behavior — a statistically grounded path toward safe, reliable autonomous systems.

Currently. My postdoctoral work continues along three threads: stochastic modeling of human social behaviors, dynamic interaction modeling (drivers and pedestrians), and Bayesian approaches to trustworthy, cognitively grounded world models. I’m always glad to hear from people working on related problems — write to me at enzozcy@gmail.com, or browse my CV.

Featured Research

Traffic Flow Theory & Simulation

Bayesian calibration of car-following models (IEEE T-ITS)
Calibrating Car-Following Models via Bayesian Dynamic Regression (ISTTT25 & TR Part C)

Social interactions for autonomous driving: A review and perspectives (Foundations and Trends® in Robotics)
Spatiotemporal learning of multivehicle interaction patterns in lane-change scenarios (IEEE T-ITS)
Interactive Car-Following: Matters but NOT Always (IEEE ITSC23)

Spatiotemporal Data & Interpretable Patterns

Markov Regime-Switching Intelligent Driver Model for Interpretable Car-Following Behavior (arXiv: 2506.14762)
Discovering dynamic patterns from spatiotemporal data with time-varying low-rank autoregression (IEEE TKDE)
Forecasting sparse movement speed of urban road networks with nonstationary temporal matrix factorization (Transportation Science)

Robust Uncertainty & Trustworthy AI

When Context Is Not Enough: Modeling Unexplained Variability in Car-Following Behavior (ISTTT26)
Active Simulation-Based Inference for Scalable Car-Following Model Calibration (arXiv: 2602.05246)

Selected Publications

Markov Regime-Switching Intelligent Driver Model for Interpretable Car-Following Behavior Chengyuan Zhang, Cathy Wu, Lijun Sun — preprint (2025) arXiv poster

Bayesian Calibration of the Intelligent Driver Model Chengyuan Zhang, Lijun Sun — IEEE T-ITS (2023) code video

Spatiotemporal Learning of Multi-Vehicle Interaction Patterns in Lane-Change Scenarios Chengyuan Zhang, Jiacheng Zhu, Wenshuo Wang, Junqiang Xi — IEEE T-ITS (2021) code demo project