Deep Reinforcement Learning with Python: With PyTorch, TensorFlow and OpenAI Gym

© Nimish Sanghi 2021

N. SanghiDeep Reinforcement Learning with Pythonhttps://doi.org/10.1007/978-1-4842-6809-4_7

7. Policy Gradient Algorithms

Nimish Sanghi¹  

(1)

Bangalore, India

 

Up to now we have been focusing on model-based and model-free methods. All the algorithms using these methods estimated the action values for a given current policy. In a second step, these estimated values were used to find a better policy by choosing the best action in a given state. These two steps were carried out in a loop again and again until no further improvement in values was observed. In this chapter, we will look at a different approach for learning optimal policies by directly operating in the policy space. We will improve the policies without explicitly learning ...