05. Q-Learning
Q-Learning
Note: One drawback of both SARSA & Q-Learning, since they are TD approaches, is that they may not converge on the global optimum when using non-linear function approximation.
Q-Learning
Note: One drawback of both SARSA & Q-Learning, since they are TD approaches, is that they may not converge on the global optimum when using non-linear function approximation.