Build a Doom AI Model with Python - Gaming Reinforcement Learning Full Course

This course builds a Python reinforcement learning agent that plays Doom. It covers VizDoom setup, OpenAI Gym environments, PPO training, testing across levels, reward shaping, and curriculum learning.

Syllabus

- Start
- Introduction
- Explainer
- CLIENT CONVERSATION 1
- Animation 1
- Tutorial Kickoff
- Getting VizDoom up and running
- CLIENT CONVERSATION 2
- Animation 2
- Creating an OpenAI Gym Environment
- CLIENT CONVERSATION 3
- Animation 3
- Setup Training Callback
- Train the RL model
- CLIENT CONVERSATION 4
- Testing the Agent
- BASIC LEVEL AI RESULTS
- CLIENT CONVERSATION 5
- Animation 4
- Changing Levels
- DEFEND CENTER LEVEL AI RESULTS
- CLIENT CONVERSATION 6
- Reward shaping
- Curriculum Learning
- DEADLY CORRIDOR LEVEL AI RESULTS
- FINAL CLIENT CALL
- Wrap up