Course Websites

Subject offerings of new and developing areas of knowledge in industrial engineering intended to augment the existing curriculum. See Class Schedule or departmental course information for topics and prerequisites. Course Information: 1 to 4 graduate hours. No professional credit. Approved for Letter and S/U grading. May be repeated in the same or separate terms if topics vary.

Prerequisite: MATH 257, MATH 415, or equivalent course on linear algebra and MATH 362, MATH 461, or equivalent course on probability. Course Description: This course will discuss the algorithm design, theoretical analysis, and simulations of dynamic programming (DP) and reinforcement learning (RL) in either finite horizon or infinite horizon, with either full observations or partial observations. Most discussions will focus on the tabular case. DP and RL with function approximations will also be introduced if time permits. This course covers foundational topics in theory of machine learning for modern use, including statistical, computational, and causal considerations in large-scale and online scenarios. We start with the classical framework of statistical learning theory and a basic probability and optimization toolkit required for understanding machine learning. We then explore two different frameworks, online/bandit learning and causal inference, highlighting the unique challenges