Can I do AI?
summary
Shows how to build agents that learn to play simple Flash games from pixels using reinforcement learning and deep neural networks, implemented with Java/Python tooling and cloud training (mentions Google Cloud and Spark/TensorFlow ecosystems). Claims models are trained from scratch and sometimes reach human-level scores (examples include runs topping ~600 points) and can reproduce AlphaGo‑style selfplay breakthroughs; demos include working agents, training pipeline, and deployment notes.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 10:10What is artificial intelligence
- 20:53AI applications and examples
- 30:44Learning resources and courses
- 40:18Neural networks basics
- 50:13Q&A and summary
