GECCO 2025 | Pareto Optimization for Subset Selection: Theories and Practical Algorithms

Pareto Optimization for Subset Selection: Theories and Practical Algorithms

Description

Subset selection aims to select a subset from a total set of items for optimizing some given objective function while satisfying some constraints. It is NP-hard, and has applications in diverse areas, including machine learning, data mining, natural language processing, computer vision, information retrieval, etc. This tutorial will introduce a promising framework, Pareto optimization, for the subset selection problem. The main idea of Pareto optimization is to transform the subset selection problem into a bi-objective optimization problem, then employ a multi-objective evolutionary algorithm to solve it, and finally return the best feasible solution w.r.t. the original subset selection problem from the generated non-dominated solution set. We will show that it achieves the best-so-far theoretical and practical performances in several applications of subset selection. We will also introduce advanced variants of Pareto optimization for large-scale, noisy and dynamic subset selection. Finally, we will briefly review the recent progress on this topic.

Organizers

Chao Qian

Chao Qian is a Professor in the School of Artificial Intelligence, Nanjing University, China. He received the BSc and PhD degrees in the Department of Computer Science and Technology from Nanjing University. After finishing his PhD in 2015, he became an Associate Researcher in the School of Computer Science and Technology, University of Science and Technology of China, until 2019, when he returned to Nanjing University as an Associate Professor. In 2024, he became a Full Professor.

His research interests include artificial intelligence, evolutionary computation, and machine learning. He has published one book “Evolutionary Learning: Advances in Theories and Algorithms”, and over 60 first/corresponding-authored papers in top-tier journals (PNAS, AIJ, ECJ, TEvC, Algorithmica, TCS) and conferences (AAAI, IJCAI, ICML, NeurIPS, ICLR). He has won the ACM GECCO 2011 Best Theory Paper Award, the IDEAL 2016 Best Paper Award, the IEEE CEC 2021 Best Student Paper Award Nomination, and the 21st Annual Humies Bronze Award. He serves on the editorial board of Artificial Intelligence Journal, Evolutionary Computation Journal, IEEE Transactions on Evolutionary Computation, IEEE Computational Intelligence Magazine, etc. He is the founding chair of IEEE Computational Intelligence Society (CIS) Task Force on Evolutionary Learning, and was also the chair of IEEE CIS Task Force on Theoretical Foundations of Bio-inspired Computation. He has regularly given tutorials and co-chaired special sessions at CEC, GECCO and PPSN, given an Early Career Spotlight Talk at IJCAI 2022, and will be a Program Co-Chair of PRICAI 2025. He has successfully developed algorithms to solve complex optimization problems (e.g., supply chain, wireless network, and chip register optimization) in Huawei, and won Huawei Spark Award twice. He is a recipient of the National Science Foundation for Excellent Young Scholars (2020) and CCF-IEEE CS Young Computer Scientist Award (2023), and has hosted a National Science and Technology Major Project.