Keke AI Competition: Solving puzzle levels in a dynamically changing mechanic space

TL;DR

The paper presents a competition framework for AI agents to solve dynamically changing puzzle levels in Baba is You, emphasizing the challenges of adapting to rule-based mechanic alterations.

Contribution

It introduces a novel competition setup and evaluation metrics for AI in a dynamic rule-based puzzle environment, with baseline results from sample agents.

Findings

Baseline tree search agents demonstrate the difficulty of the task.

The framework effectively ranks diverse AI strategies.

Dynamic mechanics significantly impact agent performance.

Abstract

The Keke AI Competition introduces an artificial agent competition for the game Baba is You - a Sokoban-like puzzle game where players can create rules that influence the mechanics of the game. Altering a rule can cause temporary or permanent effects for the rest of the level that could be part of the solution space. The nature of these dynamic rules and the deterministic aspect of the game creates a challenge for AI to adapt to a variety of mechanic combinations in order to solve a level. This paper describes the framework and evaluation metrics used to rank submitted agents and baseline results from sample tree search agents.