Reinforcement Learning Basics

---

python 3.7

gym 0.18.0

tensorflow 1.15.0

torch 1.7.1

---

Model-free (tabular setting)

1. Find Value Function

• Monte Carlo
• Temporal Difference (TD)

2. Find Optimal Policy

• Monte Carlo Control
• SARSA
• Q Learning

Deep RL (non-tabular setting)

1. Value-Based

• DQN

2. Policy-Based (policy gradient)

• REINFORCE
• PPO

3. Actor-Critic (value-based + policy-based)

• TD Actor-Critic