Welcome to the Optima Group
We are Learning and Optimization for Machine Intelligence (Optima) research group at the Department of Computer Science, City University of Hong Kong. Our major research focus is to combine ideas and techniques from evolutionary computation, traditional mathematical programming and machine learning for designing efficient metaheuristic algorithms for dealing with hard search and optimization problems in fields ranging from engineering design to e-commence and management planning.
The group has an excellent research track record. The multiobjective evolutionary optimization algorithms based on decomposition (MOEA/D) developed by us has become one of the most widely used algorithmic framework in the area of multiobjective evolutionary computation. See Research for more.
We are grateful for the generous support from University Grants Committee (UGC), Innovation and Technology Commission (ITC), National Natural Science Foundation of China (NSFC), Ministry of Science and Technology (MOST), Shenzhen Science and Technology Innovation Commission (STIC), Huawei, and CATL.
We are looking for passionate new PhD students, Postdocs, and Master students to join the team (more info) !
News
26 February 2026
🎉 Congratulations to the OptVerse-CityU team-a collaboration between our group and Huawei—including Dr. Fei Liu, Dr. Zidong Wang, Xin Chen, Qinglong Hu, Shunyu Yao, and Kefeng Zheng, a collaboration between our group and Huawei, which includes Dr. Fei LIU, Dr. Zi dong WANG, Xin CHEN, Qinglong HU, Shunyu YAO and Kefeng ZHENG for winning the 30-day CVRPLib BKS Challenge! This prestigious competition challenges teams to produce the best known solutions (BKS) for the large-scale Capacitated Vehicle Routing Problem (CVRP), and the OptVerse-CityU team succeeded in generating BKS for 51 out of 100 instances. Notably, their algorithm incorporates key components designed by the Evolution of Heuristics (EoH) system, marking a significant milestone as the first algorithm with AI-designed elements to win in this rigorous contest. Developed by our group and Huawei, the EoH system (its first version was called AEL) showcases innovative integration of evolutionary search and large language models.
25 February 2026
🎉 Our paper “Multimodal LLM-Assisted Evolutionary Search for Programmatic Control Policies” has been accepted by ICLR-26!
8 November 2025
🎉 Our paper “Beyond the Lower Bound: Bridging Regret Minimization and Best Arm Identification in Lexicographic Bandits” has been accepted by AAAI-26!
8 November 2025
🎉 Our paper “CoEvo: Continual Evolution of Symbolic Solutions Using Large Language Models” has been accepted by AAAI-26!
8 November 2025
🎉 Our paper “EoH-S: Evolution of Heuristic Set using LLMs for Automated Heuristic Design” has been accepted by AAAI-26!