OpenAI Super Mario Bros

A Reinforcement Learning implementation in the OpenAI Gym environment for Super Mario Bros, using the Proximal Policy Optimization (PPO) algorithm.

Overview

This project leverages the OpenAI Gym environment to train a reinforcement learning model to play Super Mario Bros. The PPO (Proximal Policy Optimization) algorithm is utilized to optimize the model's performance.

Key Features

• Preprocessing the Environment: Preparing the Super Mario Bros environment for efficient training.
• Simplified Movements: Reducing the complexity of Mario's movements to facilitate easier learning.
• Grayscale Frames: Converting frames to grayscale to decrease computational load.
• Vectorizing and Stacking Environments: Managing multiple frames simultaneously to enhance tracking and decision-making.