IE 598: Optimization and Learning under Uncertainty (Spring 2020)Graduate elective, University of Illinois at Urbana-Champaign
Uncertainty penetrates in every corner of data science and decision science, from data generation, model selection, system dynamics, algorithm design, all the way to prediction and decision making. This course will offer a broad overview of the modeling, theories, algorithms, and applications for the vibrant field of optimization and learning under uncertainty. Topics include stochastic optimization, robust linear/conic programs, two-stage stochastic programming , chance constraint programming, risk-averse optimization, data-driven distributionally robust optimization, multi-stage stochastic programming, and Markov decision problems. We will cover a wide range of solution methods including stochastic approximation, Monte Carlo sampling methods, variance reduction techniques, decomposition methods, convex relaxation, dynamic programming, reinforcement learning algorithms, and etc. We will also discuss their wide applications in machine learning, financial engineering, operations management, power systems, and control. Check course website here.