IE 3186 - APPROXIMATE DYNAMIC PROGRAMMING Minimum Credits: 3 Maximum Credits: 3 This course is an introduction to the theory and application of large-scale dynamic programming with emphasis on a broad spectrum of applications in finance, revenue management, and health policy. The first part of the course emphasis more on approximate dynamic programming algorithms. The second part of the course is devoted to the recent advances in reinforcement learning. Topics include Markov decision processes, dynamic programming algorithms, simulation-based algorithms, q-learning, theory and algorithms for value function approximation and policy search methods, stochastic approximation, r-max algorithm, online learning and regret minimization, and posterior sampling method. Academic Career: Graduate Course Component: Lecture Grade Component: Grad Letter Grade Click here for class schedule information.
Add to Portfolio(opens a new window)
|