Ludeme/LudiiAICompetition
Repository for the Ludii AI Competition
Provides a framework for developing general game-playing AI agents across 1,000+ board games using Ludii's forward model API, with agents implementing a standard Java "AI" abstract class to access game rules, legal moves, and state transitions. Supports three competition tracks—Kilothon (win-rate optimization against UCT), General Game Playing (unknown games), and Learning (pre-announced games)—each with agents constrained to one minute thinking time per game. Integrates Java and Python implementations through a wrapper architecture, allowing agents to query and manipulate game states without direct rule knowledge.
No commits in the last 6 months.
Stars
32
Forks
6
Language
—
License
MIT
Category
Last pushed
Jul 29, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Ludeme/LudiiAICompetition"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
tianyume/Dominoes
Dominoes is a 2D boarding game providing functionalities of player vs AI and AI training. This...
Dbz/Battleship
Battleship game against AI in python
srid/Pallanguzhi
A Tamil mancala game written in PureScript
arda-guler/HaritaWars
Mini strategy game to pit AIs against each other.
TheThirteenTeam/Domino-Monetario
Game using inteligent agents for the AI discipline.