Academics
  /  
Courses
  /  
Descriptions
  /  
Keep
IEMS 490: Special Topics: Data-driven Optimization


VIEW ALL COURSE TIMES AND SESSIONS

Prerequisites

IEMS 450-1 or a graduate level course in linear and/or convex optimization. IEMS 401 or a graduate level course in mathematical statistics.

Description

  •  This course will introduce you to the modern topic of data driven decision making under uncertainty. It is a rapidly evolving field with numerous applications in Engineering, Management and Basic Sciences. The mathematical foundations of the course involve concepts in convex analysis, risk measures, stochastic optimization, and robust optimization. This course will provide an overview of modeling frameworks for risk and data ambiguities; and algorithmic tools from stochastic and robust optimization. 

Topics

  1. Risk and Measures of Risk
  2. Modeling Ambiguity in Data
  3. Uncertainty and Ambiguity Sets
  4. Relationships between Robustness and Risk Aversion
  5. Distributionally Robust Decisions Under Uncertainty
  6. Price of Robustness
  7. Algorithms for Distributionally Robust Optimization with Continuous and Discrete Variables
  8. Utilities and Ambiguity in Preference Specification
  9. Decisions Under Risk Aversion
  10. Applications and Empirical Evidence on the Importance of Distributionally Robust Decisions