| |
Aims:
• Introduction to reinforcement learning (RL): Markov decision process, dynamic programming, Q-learning, SARSA, Actor-Critic, policy-based RL, value-based RL.
• Reinforcement learning in continuous state-action spaces. Function approximation problem.
• Reinforcement learning for robotics: mission and problems. Optimal control. Biased sampling, risk of damage, ware-out problem.
• Model-free reinforcement learning (GMMRL, PI2).
• Model-based reinforcement learning (PILCO, PI-REM).
• Approaches combining nonlinear optimal control (ILQR, MPC) and reinforcement learning.
• Introduction to deep reinforcement learning (end-to-end approaches).
|
DEEP REINFORCEMENT LEARNING FOR ROBOTS
