Gradient Descent Methods For Function Approximation

Apply Stochastic Gradient Descent (SGD) to reinforcement learning. Learn how to update function weights to minimize the mean squared value error objectively.