Overview
What this challenge is about.
Implement a small two-player imperfect-information card game (Kuhn poker or a 3-card simplified Hold'em variant). Implement CFR or CFR+ for the game and run self-play for at least 100,000 iterations. Plot exploitability over time. Build a small visualization (Streamlit or pure Jupyter) where a student can play against the trained agent at different training checkpoints to feel the learning curve. Write 2-page instructor notes explaining the algorithm in plain language plus 3 follow-up exercises students can try.
The Brief
What you'll do, and what you'll demonstrate.
Train a self-play CFR agent on a small imperfect-information card game and ship a workshop-ready demo.
Earning criteria — what you'll demonstrate
- Implement CFR or CFR+ for an imperfect-information game
- Measure exploitability as the convergence metric
- Translate self-play training into a visual learning curve students can feel
- Author instructor-grade explanatory material
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
ML Researcher
Implementing CFR from scratch and reporting honest convergence metrics is the textbook entry-level research project.
This challenge sharpens
- counterfactual-regret-minimization
- self-play
- game-theory
Research Scientist
Reproducing a published algorithm with proper exploitability measurement is the research scientist's first-week deliverable.
This challenge sharpens
- game-theory
- self-play
- counterfactual-regret-minimization
AI Product Designer
Turning a research method into a workshop-ready interactive demo is the AI product designer's bridge from algorithm to learner experience.
This challenge sharpens
- educational-content-design
- python
- reinforcement-learning