Lab Home | Phone | Search
	Center for Nonlinear Studies
	Home  ▶ People   Current   Executive Committee   Postdocs   Visitors   Students   Research   Publications  ▶ Conferences   Workshops   Sponsorship  ▶ Talks   Seminars   Postdoc Seminars Archive   Quantum Lunch   Quantum Lunch Archive   P/T Colloquia   Archive   Ulam Scholar     Postdoc Nominations  ▶ Student Requests   Student Program  ▶ Visitor Requests   Description   Past Visitors  ▶ Services   General     History of CNLS     Maps, Directions   CNLS Office   T-Division   LANL				Monday, October 14, 202410:00 AM - 11:00 AMCNLS Conference Room (TA-3, Bldg 1690) Seminar Applications of Machine Learning in Solving and Explaining Optimization Models Can LiPurdue University This talk is structured into two parts. The first part discusses the application of machine learning to enhance optimization algorithms. Combinatorial optimization and global optimization are established fields within operations research and computer science. Recently, there has been a marked increase in interest towards leveraging machine learning as a novel strategy for addressing combinatorial problems, either by serving as direct solvers or by improving the efficacy of exact solvers. We will present a new technique for solving AC optimal power flow problems with machine learning, ensuring the satisfaction of safety-critical constraints.The second part introduces OptiChat, a chatbot powered by large language models, designed to explain optimization problems. A significant obstacle to the practical deployment of optimization models is the challenge associated with helping practitioners comprehend and interpret these models. OptiChat is capable of performing a range of tasks, including diagnosing infeasibilities, conducting sensitivity analyses, providing counterfactual explanations, and responding to general inquiries from users. Host: Saif R. Kazi (T-5), Harsha Nagarajan (T-5)