Games

Explore different games and their AI solutions, organized by complexity from simple combinatorial games to complex strategic games.

Zero-Sum 2-Player Games

These games have perfect information and deterministic outcomes. They are ideal for teaching fundamental AI concepts like minimax and alpha-beta pruning.

Tic-Tac-Toe

Difficulty: Easy | Algorithm: Minimax

The simplest game, perfect for learning minimax algorithm and game tree evaluation. With perfect play from both sides, every game ends in a draw.

Checkers

Difficulty: Hard | Algorithm: Alpha-Beta Pruning

A classic game demonstrating alpha-beta pruning, move ordering optimizations, and opening books. One of the first games “solved” by computer.

Connect-4

Difficulty: Medium | Algorithm: Alpha-Beta + Transposition Tables

Learn about transposition tables and efficient move ordering. The first player has a winning strategy that’s been completely computed.

Multi-Player & Non-Zero-Sum Games

These games feature multiple players, hidden information, or probabilistic elements. They require more sophisticated AI techniques like Monte Carlo Tree Search and probabilistic reasoning.

Catan

Difficulty: Hard | Algorithm: Monte Carlo Tree Search

A complex strategy game demonstrating MCTS for games with hidden information and high branching factors. Players compete for resources and victory points.

Wingspan

Difficulty: Hard | Algorithm: Bayesian Inference, Belief State Tracking

A probabilistic card game requiring reasoning about uncertainty and player modeling. AI must manage hidden deck states and opponent card probabilities.

Dune Imperium

Difficulty: Hard | Algorithm: Multi-Objective Optimization

A multi-objective strategy game with complex interactions, simultaneous decisions, and political intrigue. AI balances military, economic, and political goals.

Algorithm Reference

Not sure which algorithm applies to which game? Check out the Algorithm Reference for detailed explanations, complexity analysis, and pseudocode for:

Next Steps