LANS Seminar

Seminar Title: Reliable and Adaptive Stochastic Optimization in the Face of Messy Data

Speaker: Miaolan Xie, Postdoctoral Researcher, University of Chicago, As of July 2025, Assistant Professor in Industrial Engineering, Purdue

Date/Time: Thursday, April 3, 2025/ 2:30 PM – 3:30 PM (In-Person)
Location: See Meeting URL on the cels-seminars website which will require an Argonne login.

Description:  Solving real-world stochastic optimization problems presents two key challenges: the messiness of real-world data, which can be noisy, biased, or corrupted due to factors like outliers, distribution shifts, and even adversarial attacks; and the laborious, time-intensive requirement of manually tuning step sizes in many existing algorithms. I will introduce a simple adaptive optimization framework that avoids the need for manual step size tuning by adaptively adjusting the step size in each iteration based on the progress of the algorithm. To address the issue of messy data, the algorithm only assumes access to function-related information through probabilistic oracles, which may be biased and corrupted. This framework is very general, encompassing a wide range of algorithms for both unconstrained and constrained optimization. It is applicable to multiple problem settings, such as expected loss minimization in machine learning, simulation optimization, and derivative-free optimization. Under reasonable conditions on the oracles, we provide an analytical framework that bounds the iteration complexity with high probability for settings with highly noisy and corrupted gradients.

Bio: Miaolan Xie will join Purdue University in July 2025 as an Assistant Professor in Industrial Engineering. She is currently a postdoctoral researcher at the University of Chicago, working with Dan Adelman. She obtained her PhD in Operations Research and Information Engineering from Cornell University, advised by Katya Scheinberg. Her research lies in the intersection of machine learning, optimization, utilizing tools from statistics and stochastic processes, with applications in data science and healthcare. Miaolan has received several awards for her research, including Second Place in the 2023 Student Paper Prize awarded by the INFORMS Optimization Society and Second Place in the Flash Talk Competition at the YinzOR Student Conference. In 2022, she was awarded NSF Research Internship funding to work as a Givens Associate in the Mathematics and Computer Science Division at Argonne National Laboratory with Stefan Wild and Matt Menickelly. Prior to her doctoral studies, Miaolan worked as a data scientist at Alibaba Group on the retail supply chain platform team.

Please note that the meeting URL for this event can be seen on the cels-seminars website which requires an Argonne login.