Deep Reinforcement Learning for Cartpole Environment

Decription

In this project, we implement a policy gradient method using deep neural networks to train our agent in cartpole environment from the gym. For this implementation, we’ve used the Tensorflow framework.

Environment

In cartpole Environment, our agent just has two actions to do:

• +1: push cart to the right
• 0: push cart to the left our goal is to keep the balance of this cart.

Training

We train our agent for 1000 episodes