{"id":3887,"date":"2025-10-13T15:40:37","date_gmt":"2025-10-13T20:40:37","guid":{"rendered":"https:\/\/wordpress.cels.anl.gov\/lans-seminars\/?post_type=tribe_events&p=3887"},"modified":"2025-10-13T15:41:38","modified_gmt":"2025-10-13T20:41:38","slug":"3887","status":"publish","type":"tribe_events","link":"https:\/\/wordpress.cels.anl.gov\/lans-seminars\/event\/3887\/","title":{"rendered":"LANS Seminar"},"content":{"rendered":"

Seminar Title<\/strong>: Distributional Control as a Representation Learning Machine<\/p>\n

Speaker<\/strong>: Jr-Shin Li, Newton R. and Sarah Louisa Glasgow Wilson Professor,\u00a0 Department of Electrical and Systems Engineering, Washington University in St. Louis<\/p>\n

Date:<\/strong> Thursday, October 16, 2025<\/p>\n

Time:<\/strong>\u00a02:30 PM-3:30 PM (In-Person)<\/p>\n

Location:<\/strong>\u00a0Hybrid, Bldg. 240, Conference Room 4301<\/p>\n

Host:<\/strong> Krishnan Raghavan<\/p>\n

\n

Description<\/strong>: The construction of interpretable representation models for learning flows of probability distributions has become a rapidly growing research area in machine learning, opening new research avenues and offering unique insights into established fields such as image processing and robotics. In this talk, we will introduce ensemble control systems (ECS) as interpretable machines for learning and generation of probability flows. The ECS model is parameterized by heterogeneous control systems with control inputs acting as time-dependent trainable parameters. The heterogeneous dynamics and time-dependent parameters significantly enhance the model\u2019s capabilities, making it highly effective. To leverage these strengths, we will present a moment kernelization approach that generates reduced kernel representations of ECS within a reproducing kernel Hilbert space, enabling efficient training. The significant advantages of the ECS model will be demonstrated through various image restoration tasks, alongside a comparison with baseline flow matching-based image processing models.<\/p>\n

Bio: <\/span><\/b>Dr.\u00a0Jr-Shin\u00a0Li\u00a0is Newton R. and Sarah Louisa Glasgow Wilson Professor in the Department of Electrical and Systems Engineering at Washington University in St. Louis. He also holds joint appointments in the Division of Computational \\& Data Sciences (DCDS) and the Division of Biology \\& Biomedical Sciences (DBBS). Dr.\u00a0Li\u00a0received his B.S. and M.S. degrees from National Taiwan University and his Ph.D. in Applied Mathematics from Harvard University in 2006. His research interests encompass systems, computational, and data sciences with applications in biology, neuroscience, quantum control, brain medicine, public health, complex networks, and machine learning. Dr.\u00a0Li\u00a0is a recipient of the NSF CAREER Award (2008) and the AFOSR Young Investigator Award (2010). He has served as Associate Editor for the SIAM Journal on Control and Optimization (SICON) and the IEEE Transactions on Control Systems Technology (TCST). Currently, he is an Editorial Member of Nature Scientific Reports. Dr.\u00a0Li\u00a0is an IEEE Fellow and a founder and co-Chair of the IEEE Technical Committee on Quantum Computing, Systems, and Control since 2024.<\/span><\/p>\n

Please note that the meeting URL for this event can be seen on the cels-seminars website which requires an Argonne login.<\/em><\/p>\n

\n

See all upcoming talks at\u00a0https:\/\/www.anl.gov\/mcs\/lans-seminars<\/a><\/p>\n<\/div>\n<\/div>\n

\n