Basic Conceptions of Reinforcement learning

19 March 2017 - Beijing

Dynamics of Markov Desion-making Process(MDP):

Model-based Reinforcement Learning:

Models assuming that the dynamics of MDP are known, such as dynamic programming

Model-free Reinforcement Learning:

Models like Monte-Carlo evaluation or temporal-difference learning, learning directly from experience and do not assume any knowledge of the environment’s dynamics

references

  1. David Silver’s thesis
Written on March 19, 2017