Co Evolution
Co-evolution studies the intertwined evolution of interacting systems, aiming to understand how reciprocal selective pressures shape their development and adaptation. Current research focuses on developing and applying co-evolutionary algorithms, including those leveraging large language models and phylogenetic analysis, to optimize diverse systems, from enzyme design and robot-human collaboration to multi-agent communication and structural engineering. These advancements offer improved efficiency and problem-solving capabilities in various fields, while also providing insights into fundamental biological and computational processes.
Papers
Unlocking the Potential of Weakly Labeled Data: A Co-Evolutionary Learning Framework for Abnormality Detection and Report Generation
Jinghan Sun, Dong Wei, Zhe Xu, Donghuan Lu, Hong Liu, Hong Wang, Sotirios A. Tsaftaris, Steven McDonagh, Yefeng Zheng, Liansheng Wang
Efficient Language-instructed Skill Acquisition via Reward-Policy Co-Evolution
Changxin Huang, Yanbin Chang, Junfan Lin, Junyang Liang, Runhao Zeng, Jianqiang Li