Markov Decision Processes

• Some of this material is also covered in ML for Trading in Reinforcement Learning
• The concept is also covered in AI for Robotics in Motion Planning

It does not tell you the plan of what sequence of action to take, it takes you what action to take given that you land in a particular state.

The top section defines the Problem whereas the policy section defines the Solution. We could use policy-based or value-based solvers to get from Problem to Solution.