Adversarial Coevolutionary Illumination with Generational Adversarial MAP-Elites
Adversarial Coevolutionary Illumination with Generational Adversarial MAP-Elites
Unlike traditional optimization algorithms focusing on finding a single optimal solution, Quality-Diversity (QD) algorithms illuminate a search space by finding high-performing solutions that cover a specified behavior space. However, tackling adversarial problems is more challenging due to the behavioral interdependence between opposing sides. Most applications of QD algorithms to these problems evolve only one side, thus reducing illumination coverage. In this paper, we propose a new QD algorithm, Generational Adversarial MAP-Elites (GAME), which coevolves solutions by alternating sides through a sequence of generations. Combining GAME with vision embedding models enables the algorithm to directly work from videos of behaviors instead of handcrafted descriptors. Some key findings are that (1) emerging evolutionary dynamics sometimes resemble an arms race, (2) starting each generation from scratch increases open-endedness, and (3) keeping neutral mutations preserves stepping stones that seem necessary to reach the highest performance. In conclusion, the results demonstrate that GAME can successfully illuminate an adversarial multi-agent game, opening up interesting future directions in understanding the emergence of open-ended coevolution.
Timothée Anne、Noah Syrkis、Meriem Elhosni、Florian Turati、Franck Legendre、Alain Jaquier、Sebastian Risi
自然科学研究方法系统科学、系统技术信息科学、信息技术
Timothée Anne,Noah Syrkis,Meriem Elhosni,Florian Turati,Franck Legendre,Alain Jaquier,Sebastian Risi.Adversarial Coevolutionary Illumination with Generational Adversarial MAP-Elites[EB/OL].(2025-05-10)[2025-06-03].https://arxiv.org/abs/2505.06617.点此复制
评论