Learning Car-Following Behaviors Using Bayesian Matrix Normal Mixture Regression

arXiv:2404.16023v1 Announce Type: cross
Abstract: Learning and understanding car-following (CF) behaviors are crucial for microscopic traffic simulation. Traditional CF models, though simple, often lack generalization capabilities, while many data-driven methods, despite their robustness, operate as "black boxes" with limited interpretability. To bridge this gap, this work introduces a Bayesian Matrix Normal Mixture Regression (MNMR) model that simultaneously captures feature correlations and temporal dynamics inherent in CF behaviors. This approach is distinguished by its separate learning of row and column covariance …

abstract arxiv bayesian black boxes bridge capabilities car cs.lg data data-driven gap interpretability matrix normal regression robustness simple simulation stat.ap traffic type understanding while work