01 Basic Concepts

• Concepts:
  • state
  • action
  • reward
  • return
  • episode
  • policy
• Grid-world example
• MDP

A grid-world example

• how to define “good”? Avoid forbidden cells, detours, or boundary

Summary

By a grid-world examples, learn the key concepts:

• state
• action
• state transtion, state transition probability
• reward, reward probability
• trajectory, episode, return, discounted return
• Markov Decision Process

Later

Fundermental concepts, widely used later

next→02 Bellman Equation