Interpreting Multi-objective Evolutionary Algorithms via Sokoban Level Generation
Qingquan Zhang, Yuchen Li, Yuhang Lin, Handing Wang, Jialin Liu
TL;DR
The paper tackles the challenge of understanding multi-objective evolutionary algorithms (MOEAs) by providing an interactive, web-based platform. It uses Sokoban level generation to demonstrate balancing $f_{emp}$ and $f_{div}$ with the Two_Arch2 algorithm. Contributions include adapting Two_Arch2 to Sokoban, implementing real-time visualizations, and step-by-step guidance for learners. The platform clarifies CA and DA roles and shows Pareto-front relationships, making MOEAs tangible for education and PCG practice. Public accessibility via a web interface supports researchers, students, and educators in exploring MOEAs and PCG.
Abstract
This paper presents an interactive platform to interpret multi-objective evolutionary algorithms. Sokoban level generation is selected as a showcase for its widespread use in procedural content generation. By balancing the emptiness and spatial diversity of Sokoban levels, we illustrate the improved two-archive algorithm, Two_Arch2, a well-known multi-objective evolutionary algorithm. Our web-based platform integrates Two_Arch2 into an interface that visually and interactively demonstrates the evolutionary process in real-time. Designed to bridge theoretical optimisation strategies with practical game generation applications, the interface is also accessible to both researchers and beginners to multi-objective evolutionary algorithms or procedural content generation on a website. Through dynamic visualisations and interactive gameplay demonstrations, this web-based platform also has potential as an educational tool.