报告人：李娜 教授
报告题目：Inverse Optimal Control for Linear Quadratic Problem with Poisson Jumps: Model-Free Inverse Reinforcement Learning Approaches
报告时间：2026年10月24日（周六）上午9：00
报告地点：云龙校区智华楼201报告厅
主办单位：数学与统计学院、数学研究院、科学技术研究院
报告人简介：
李娜，大连理工大学数学科学学院教授，博士生导师，入选国家高层次青年人才计划，国务院政府特殊津贴专家，省政府特殊津贴专家，首届省级科学技术青年奖获得者，首届山东数学会青年数学奖获得者，山东省教育系统优秀共产党员；兼任教育部高校教指委委员、省科协第十届委员会委员、中国自动化学会TCCT随机系统控制委员会委员、山东数学会常务理事、山东省青年科学家协会理事、山东省大数据研究会理事等。近年来，在 随机最优控制、随机微分博弈、强化学习等方向取得了一系列研究成果，发表在《SIAM Journal on Control and Optimization》、《Automatica》、《IEEETransactions on Automatic Control》等国际著名学术期刊；先后主持国家自然科学基金项目、省自然科学基金项目多项；曾获省级教学成果奖一等奖1项、二等奖4项，省高等学校科学技术奖二等奖1项等；主持教育部供需对接就业育人项目1项、产学合作协同育人项目2项、省研究生教育优质课程建设项目1项、省本科教学改革研究重点项目1项等。
报告摘要：
This work addresses an inverse optimal control (IOC) problem for stochastic linear systems subject to both Brownian motion and Poisson jumps, using an inverse reinforcement learning (IRL) framework. Given a target feedback gain from an expert, the objective is to identify an equivalent cost functional—specifically, the set of all cost weights—that yields this same gain. To solve this problem when system dynamics are unknown, we propose two model-free, off-policy IRL algorithms that operate entirely from data, circumventing the need to solve the generalized algebraic Riccati equation or compute the cost weights analytically. The first is an inverse Q-learning algorithm that constructs data-driven equations from ex- pert demonstrations to compute the Q-function matrix, with equivalent cost weights updated algebraically and without requiring additional trajectory data. The second is a model-free off-policy inverse policy iteration algorithm that leverages data collected under an initial stabilizing policy, offering a complementary approach suited to different data availability scenarios. Crucially, by decoupling the data-collection behavior policies from the policies being iteratively updated, both algorithms can learn equivalent cost weights from sufficiently excited trajectories without identifying the system dynamics or jump intensity. Numerical simulations validate the effectiveness of the proposed methods.
