Skip to content

kennethvuongcode/Space-Invaders-AI

Repository files navigation

Space Invaders AI

In this project, I developed an AI agent capable of playing the classic arcade game Space Invaders. The game environment was created using PyGame, and the AI leverages a Deep Q-Learning approach implemented with PyTorch. This project served as a practical introduction to coding and reinforcement learning concepts.

Key Components:

  • Game Environment: Utilized PyGame to replicate the Space Invaders game, including player controls, enemy behaviors, and game physics.
  • Neural Network: Designed a neural network using PyTorch to approximate the Q-value function, enabling the agent to predict the expected rewards of actions in given states.
  • Deep Q-Learning: Implemented a Deep Q-Learning algorithm where the agent learns optimal actions through exploration and exploitation, updating its knowledge based on game outcomes.
  • Training Pipeline: Developed a training loop that allows the agent to play multiple game episodes, learn from its experiences, and improve its performance over time.

Skills Acquired:

  • Game Development: Gained experience in developing interactive game environments using PyGame.
  • Reinforcement Learning: Learned to implement and train agents using Deep Q-Learning, a foundational reinforcement learning algorithm.
  • Neural Network Design: Acquired practical skills in designing and training neural networks with PyTorch.

This project provided a hands-on experience in combining game development with artificial intelligence, resulting in an AI agent capable of learning and playing Space Invaders autonomously. It also laid the groundwork for future explorations into more complex reinforcement learning scenarios and AI-driven game development projects.

About

Custom developed Space Invaders AI using PyGame and PyTorch. Fun project!

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages