北航经管学院“至信论坛”系列讲座
讲座题目：Representative Agents for Traffic Modeling: From Fully LLM-Driven to LLM-Guided Learning
讲座时间：2025.12.25（周四）10:00-11:30
讲座地点：新主楼A1148
讲座嘉宾：李加阳 助理教授，香港大学
讲座嘉宾 简介
Jiayang Li (李加阳) is an Assistant Professor in the Department of Data and Systems Engineering and a Fellow of the Institute of Transport Studies at the University of Hong Kong. Prof. Li received his Ph.D. in Transportation from Northwestern University in 2024 and his B.S. in Mathematics from Tsinghua University in 2019. His research aims to integrate optimization, game theory, and machine learning to address operations research challenges, with a particular focus on transportation and mobility systems. His work has been accepted for publication in leading journals and conferences in diverse fields, including Transportation Science, Transportation Research Part B, Operations Research, ISTTT, NeurIPS, and ICML.
讲座概要
Large language models (LLMs) are increasingly used as behavioral proxies for self-interested travelers in agent-based traffic models. Although more flexible and generalizable than conventional models, the practical use of these approaches remains limited by scalability due to the cost of calling one LLM for every traveler. Moreover, it has been found that LLM agents often produce unstable day-to-day dynamics. To address these challenges, we propose to model each homogeneous traveler group facing the same decision context with a single representative LLM agent who behaves like the population's average, maintaining and updating a mixed strategy over routes that coincides with the group's aggregate flow proportions. Each day, the LLM reviews the travel experience and flags routes with positive reinforcement that they hope to use more often, and an interpretable update rule then converts this judgment into strategy adjustments using a tunable (progressively decaying) step size. The representative-agent design improves scalability, while the separation of reasoning stabilizes learning. In classic traffic assignment settings, we find that the proposed approach converges rapidly to the user equilibrium. In richer settings with income heterogeneity, multi-criteria costs, and multi-modal choices, the generated dynamics remain stable and interpretable. Moreover, it reproduces behavioral patterns well-documented in psychology and economics, for instance, the decoy effect in toll and non-toll road selection and the greater willingness-to-pay for convenience among higher-income travelers.
