The task of this project is to master classic Atari games using Deep Reinforcement Learning (DRL). The challenge lies in training an AI agent to achieve high scores and optimal performance in games such as Space Invaders, Pacman, and cartpole.
In this project, we have tackled the problem by implementing state-of-the-art DRL algorithms, specifically Deep Q-Networks (DQN), to train AI agents to play Atari games. We have utilized the Stable Baselines3 library, which provides powerful tools for implementing various reinforcement learning algorithms. Our approach involves training the agents to maximize cumulative rewards by learning effective strategies and decision-making processes.
To install the project, follow these steps:
Clone the repository to your local machine. Install the required dependencies by running pip install -r requirements.txt. Ensure that you have the necessary environment set up for running the code, including Python 3.7 or above.
To use the project, follow these instructions:
Navigate to the project directory. Run the main script to start training the AI agent: Monitor the training progress and performance metrics as the agent learns to play the Atari games. Evaluate the trained agent by running the evaluation script: Analyze the evaluation results to assess the agent's performance and effectiveness in mastering the games.
./my_project argument1 argument2
deniran_o
Made at Qwasar SV -- Software Engineering School <img alt='Qwasar SV -- Software Engineering School's Logo' src='https://storage.googleapis.com/qwasar-public/qwasar-logo_50x50.png' width='20px' />