The notebook contents implementation of solution of OpenAI gym Acrobot-v1 problem.
This project aims to employ deep reinforcement learning to train an agent to play the Acrobot game in the OpenAI Gym environment. The Acrobot game is a physics-based game in which a two-link pendulum must be controlled to swing up and balance at the top. This task is challenging because it requires the agent to understand intricate dynamics and devise a strategy to keep the pendulum balanced. The project utilizes the Deep Q-Network (DQN) algorithm, a well-known method for training agents in environments with extensive state spaces.
Results for the first and last episodes are saved into .mp4 files; Learning curve for score is saved into Acrobot_score.png
