Dr.\u00a0Zhenzhen\u00a0Yan is an assistant professor at School of Physical and Mathematical Sciences, Nanyang Technological University. She joined SPMS since 2018. Before that, she received her PhD in Management Science from the National University of Singapore, and her BSc and MSc in Management Science, Operations Research from the National University of Defense and Technology in China. Her research interests mainly focus on the interplay between optimization and data analytics. Her first line of research is to solve various operations management problems and engineering problems from the distributionally robust perspective, including supply chain design and operations, and healthcare operations. The second line is to develop data-driven optimization approaches with applications to e-commerce operations and resource allocation. Her work has been published in leading operations management journals including\u00a0Management Science,\u00a0Operations Research,\u00a0MSOM and POMS, and top AI conferences including Neurips, UAI and AAAI.\u00a0Her work has\u00a0received media coverage in various\u00a0outlets including the\u00a0Straits Times and ScienceDaily etc. She currently serves as an Associate Editor of\u00a0Decision Sciences.<\/p>\n"}],"event_agenda":false,"event_photo_gallery":false,"event_presentations":false,"event_custom_heading":[{"event_custom_title":"Abstract","event_custom_details":"
We study an edge-weighted online stochastic\u00a0Generalized Assignment Problem\u00a0with\u00a0unknown\u00a0Poisson\u00a0arrivals. We provide a sample-based multi-phase\u00a0algorithm by utilizing\u00a0both pre-existing offline data (named historical data) and sequentially revealed online data. The developed\u00a0algorithm employs the concept of exploration-exploitation to\u00a0dynamically learn the arrival rate and optimize\u00a0the allocation decision. We establish its parametric performance guarantee measured by a competitive ratio.\u00a0We further provide a guideline on\u00a0fine tuning the parameters under different sizes of historical data based on\u00a0the established parametric form. By analyzing a special case which is a classical online weighted matching\u00a0problem, we also provide a novel insight on how the historical data\u2019s quantity and quality (measured by the\u00a0number of underrepresented agents in the data) affect the trade-off between\u00a0exploration and exploitation in\u00a0online algorithms and their performance. Finally, we demonstrate the effectiveness of our algorithms numerically.<\/p>\n"}],"event_enquiry_details":{"event_enq_full_name":"","event_enq_department":"","event_enq_email":"","event_enq_telephone":"","event_enq_website":""}}]