Artificial Intelligence · CSCI 3202

Mancala AI Agent

A Mancala-playing AI built from scratch in Python (game engine, random baseline, and search agents), benchmarked head-to-head over 100-game tournaments. Minimax with alpha-beta pruning plus a custom heuristic lifts win rate to 99% at depth 5.

Python NumPy Matplotlib

99% win rate at depth 5 with the new heuristic

~3.9× speedup from alpha-beta pruning at identical win rates

10 deepest search depth made feasible by pruning

100 games per tournament for benchmarking each agent

How it's built

The project is structured as four layers, each independently testable:

Game engine. A Mancala board with move legality, stone capture, end-of-game detection, and turn alternation, built from scratch so every search agent talks to the same rules.
Random baseline. Picks any legal move uniformly: the floor every smarter agent has to beat by a wide margin to count.
Minimax + alpha-beta search. Identical move choices to plain minimax with a fraction of the node visits, making deeper plies practical.
Custom heuristic evaluation. A linear combination of side-store difference and end-of-game stone-sweep credit, tuned through tournament play.

The interesting part: pruning vs. heuristic

Plain minimax at depth 10 was projected to take roughly two hours per game, clearly not playable. Alpha-beta pruning kept the same move choices but cut node visits by enough to make depth-10 search feasible in seconds, a ~3.9× speedup.

Even with deeper search, win rate against the baseline plateaued around 96%. The last gains didn't come from more depth. They came from teaching the heuristic to value end-of-game stone sweeps, which are decisive late in a game but invisible to a naive score-difference function. That single change pushed win rate to 99% at depth 5.

Key insight

Past a certain depth, marginal wins come from heuristic design, not raw search. Crediting end-of-game sweeps was a tiny code change that delivered the final 3%, bigger than any extra ply of search had.

What I took away

Alpha-beta isn't just a speedup. It's what makes deeper, otherwise-impractical search affordable in the first place.
Heuristic design eventually dominates search depth. A small domain-aware tweak beat several plies of additional lookahead.
Tournament play is the honest benchmark. Single games hide variance; 100-game tournaments make small heuristic edges visible.