Skip to content

Fifth year

Reinforcement and Imitation Learning

Reinforcement and imitation learning: Markov decision processes, Bellman equations, Q-learning, DQN, policy gradient and PPO, behaviour cloning and DAgger, Gymnasium, Stable-Baselines3 and PufferLib, simulation and Gaussian splatting, applications.

9 articles

Course syllabus

9 topics, in course order

100%complete

9 of 9 topics have a complete lesson

  1. 1 Markov decision processes
  2. 2 Bellman equations
  3. 3 Temporal differences, SARSA and Q-learning
  4. 4 Deep Q-Network
  5. 5 Policy gradient and PPO
  6. 6 Imitation learning
  7. 7 Libraries: Gymnasium, Stable-Baselines3, PufferLib
  8. 8 Simulation and Gaussian splatting
  9. 9 Applications: robotics, driving, RLHF