The course covers graduate level dynamic optimization problems that can be tackled by computational approach. Assuming prior knowledge in MCMC and Bayesian applications, temporal difference modeling and functional approximations are the central methods in the course.

Models that will be covered are

• Segmented recommendation
  – Gamma-Poison
  – Tensor Factorization
• Multi-agent reinforcement learning
  – Multi-agent Nash Q-learning
  – N-period / N-player optimizations
• Project
  – AWS DeepRacer

Other topics to be added later.