Dr. Fengqi You
Roxanne E. and Michael J. Zak Professor in Energy Systems Engineering
Smith School of Chemical and Biomolecular Engineering,
Cornell University

Part of the Fall 2019 Graduate Seminar Series

When Machine Learning Meets Robust Optimization: Data-driven Adaptive Robust Optimization Models, Algorithms & Applications

Abstract:

Data-driven decision-making is an emerging paradigm that integrates model-based and data-driven systems for optimization under uncertainty. This organic integration of machine learning and mathematical programming leads to fundamentally more powerful and efficient data-driven optimization frameworks that close the loop between data analytics and decision support.  In this presentation, we will introduce a data-driven adaptive robust optimization framework, which leverages Bayesian nonparametric methods to derive flexible data-driven uncertainty sets that were previously intractable for conventional robust optimization. The machine learning model is seamlessly integrated with adaptive robust optimization approach through a novel multi-level optimization framework, which explicitly accounts for the correlation, asymmetry and multimode of uncertainty data and automatically adjusts its complexity based on the data structure and complexity. The data-driven adaptive robust optimization framework is further extended to systematically and automatically handle labeled multi-class uncertainty data through a stochastic robust optimization approach with a bi-level structure. Tailored decomposition algorithms are further developed to solve the resulting multi-level optimization problem efficiently. Applications to process manufacturing, energy systems and digital agriculture will be presented to demonstrate the applicability of the proposed frameworks.

Biography:

Fengqi You is the Roxanne E. and Michael J. Zak Professor in Smith School of Chemical and Biomolecular Engineering at Cornell University. He also serves as Chair of Cornell Systems Engineering PhD Studies and Associate Director of Cornell Energy Systems Institute. His research focuses on novel computational models, optimization algorithms, statistical machine learning methods, and multi-scale systems analytics tools for smart manufacturing, digital agriculture, energy systems, and sustainability. His recent awards (in the past three years) include National Science Foundation CAREER Award (2016), AIChE Environmental Division Early Career Award (2017), AIChE Sustainable Engineering Research Excellence Award (2017), Computing and Systems Technology (CAST) Outstanding Young Researcher Award from AIChE (2018), Cornell Engineering Research Excellence Award (2018), ACS Sustainable Chemistry & Engineering Lectureship Award (2018), and AIChE Excellence in Process Development Research Award (2019), as well as a number of best paper awards. For more information about his research group: www.peese.org