This course is an introduction to the theory and application of large-scale dynamic programming.
FREE
This course includes
Hours of videos
583 years, 3 months
Units & Quizzes
21
Unlimited Lifetime access
Access on mobile app
Certificate of Completion
Topics include Markov decision processes, dynamic programming algorithms, simulation-based algorithms, theory and algorithms for value function approximation, and policy search methods. The course examines games and applications in areas such as dynamic resource allocation, finance and queueing networks.
Course Currilcum
- Markov Decision Processes Unlimited
- Gauss-Seidel Value Iteration Unlimited
- Asynchronous Policy Iteration Unlimited
- Blackwell Optimality Unlimited
- Average-Cost Problems Computational Methods Unlimited
- Real-Time Value Iteration Unlimited
- Q-Learning Stochastic Approximations Unlimited
- Convergence of Q-Learning Unlimited
- Exploration versus Exploitation: The Complexity of Reinforcement Learning Unlimited
- Approximation Architectures Unlimited
- Model Selection and Complexity Unlimited
- Performance Bounds Unlimited
- Temporal-Difference Learning with Value Function Approximation Unlimited
- Temporal-Difference Learning with Value Function Approximation (cont.) Unlimited
- General Control Problems Unlimited
- Approximate Linear Programming Unlimited
- Approximate Linear Programming (cont.) Unlimited
- Efficient Solutions for Approximate Linear Programming Unlimited
- Efficient Solutions for Approximate Linear Programming: Factored MDPs Unlimited
- Policy Search Methods Unlimited
- Policy Search Methods (cont.) Unlimited
