20 items
Sampling From a Stochastic Policy
Q-Table Initialization and Indexing
States, Actions, and Rewards Defined
Policy: Mapping States to Actions
Agent, Environment, and Interaction Loop
Real-World Applications of RL
Reward Shaping and Reward Design
The Credit Assignment Problem
Episodic vs Continuing Tasks
On-Policy vs Off-Policy Learning
Temporal Difference Learning Concept
Q-Learning as a Basic RL Algorithm
Model-Based vs Model-Free RL
Markov Decision Process: The RL Framework
Epsilon-Greedy Action Selection
Exploration vs Exploitation Tradeoff
Value Functions: State and Action (Q)
Discount Factor in Reinforcement Learning
Reward Signal vs Long-Term Return
A policy is the agent's strategy — a rule that maps each state to an action to take.
