14. Distributional Reinforcement Learning

14. Distributional Reinforcement Learning

• 14.1. Why Distributional Reinforcement Learning?
• 14.2. Categorical DQN
  • 14.2.1. Predicting Value Distribution
  • 14.2.2. Selecting Action Based on the Value Distribution
  • 14.2.3. Training the Categorical DQN
  • 14.2.4. Projection Step
  • 14.2.5. Putting it all Rogether
  • 14.2.6. Algorithm - Categorical DQN
• 14.3. Playing Atari games using Categorical DQN
• 14.4. Quantile Regression DQN
• 14.5. Math Essentials
  • 14.5.1. Quantile
  • 14.5.2. Inverse CDF (Quantile function)
• 14.6. Understanding QR-DQN
  • 14.6.1. Action Selection
  • 14.6.2. Loss Function
• 14.7. Distributed Distributional DDPG
  • 14.7.1 Critic Network
  • 14.7.2. Actor Network
  • 14.7.3. Algorithm - D4PG