GRADUATE SEMINAR ON INTERACTIVE MACHINE LEARNING

Credit Hours: 1-3 · Graduate

Many applications of machine learning involve humans in the loop (e.g., the programmer implementing the algorithm, the domain expert specifying the features/labels, or the end user making decisions using the learned model). This course is a discussion-based seminar focusing on the design, analysis, and evaluation of machine learning techniques with explicit emphasis on the human(s) in the loop. Topics include reinforcement learning with human teachers, active learning, interpretability, learning beyond labels, and human-in-the-loop Bayesian inference. Recommended Prerequisite(s): COMP 382 and STAT 315 / DSCI 301 and CAAM 335 Repeatable for Credit.