会议介绍
GECCO(Proceedings of the Genetic and Evolutionary Computation Conference)是进化计算领域最大的同行评审会议，也是计算机协会（ACM）遗传与进化计算特别兴趣组（SIGEVO）的主要会议（俗称盛会型）。CORE2021排名为A，CCF分类为C。
GECCO2024共提交full paper495篇，录用178篇，录取率约为36%，比去年35%上升1%。历年录取率如下：
GECCO的每个Track都有自己的最佳论文。此外，还有Student Workshop的最佳论文。某些Track可能出现合并然后决出最佳论文的情况。以下给出最佳论文列表。论文下载见原文链接。
最佳论文
CS + L4EC Tracks
赛道：Complex Systems + Learning for Evolutionary Computation
题目：Quality with Just Enough Diversity in Evolutionary Policy Search
作者：Paul Templier (University of Toulouse, ISAE-SUPAERO), Luca Grillotti (Imperial College London), Emmanuel Rachelson (University of Toulouse, ISAE-SUPAERO), Dennis George Wilson (University of Toulouse, ISAE-SUPAERO), Antoine Cully (Imperial College London)
关键词：演化策略、策略搜索、质量多样性优化
ECOM Track
赛道：Evolutionary Combinatorial Optimization and Metaheuristics
题目：Letting a Large Neighborhood Search for an Electric Dial-A-Ride Problem Fly: On-The-Fly Charging Station Insertion
作者：Maria Bresich (Vienna University of Technology), Günther Raidl (Vienna University of Technology), Steffen Limmer (Honda Research Institute Europe GmbH)
关键词：调度问题、领域搜索
EML Track
赛道：Evolutionary Machine Learning
题目：Survival-LCS: A Rule-Based Machine Learning Approach to Survival Analysis
作者：Alexa A. Woodward (Optum Life Sciences), Harsh Bandhey (Cedars-Sinai Medical Center), Jason H. Moore (Cedars-Sinai Medical Center), Ryan J. Urbanowicz (Cedars-SiCnai Medical Center)
关键词：生存分析、遗传算法
EMO Track
赛道：Evolutionary Multiobjective Optimization
题目：Analysis of Real-World Constrained Multi-Objective Problems and Performance Comparison of Multi-Objective Algorithms
作者：Yang Nan (Southern University of Science and Technology), Hisao Ishibuchi (Southern University of Science and Technology), Tianye Shu (Southern University of Science and Technology), Ke Shang (Southern University of Science and Technology)
关键词：约束多目标优化问题、真实场景和理论场景的差异
ENUM + SI Tracks
赛道：Evolutionary Numerical Optimization + Swarm Intelligence
题目：Markov Chain-based Optimization Time Analysis of Bivalent Ant Colony Optimization for Sorting and LeadingOnes
作者：Matthias Kergaßner (University of Erlangen-Nuremberg), Oliver Keszocze (University of Erlangen-Nuremberg), Rolf Wanka (University of Erlangen-Nuremberg)
关键词：蚁群算法的运行时间复杂度分析、马尔可夫链
GA + THEORY Track
赛道：Genetic Algorithms + Theory
题目：Evolutionary Computation Meets Graph Drawing: Runtime Analysis for Crossing Minimisation on Layered Graph Drawings
作者：Jakob Baumann (University of Passau), Ignaz Rutter (University of Passau), Dirk Sudholt (University of Passau)
关键词：单侧二分交叉最小化问题、运行时间分析
GECH + NE Track
赛道：General Evolutionary Computation and Hybrids + Neuroevolution
题目：Tensorized NeuroEvolution of Augmenting Topologies for GPU Acceleration
作者：Lishuang Wang (Southern University of Science and Technology), Mengfei Zhao (Southern University of Science and Technology), Enyu Liu (Southern University of Science and Technology), Kebin Sun (Southern University of Science and Technology), Ran Cheng (Southern University of Science and Technology)
关键词：增强拓扑神经进化、GPU加速
GP Track
赛道：Genetic Programming
题目：Learning Traffic Signal Control via Genetic Programming
作者：Xiao-Cheng Liao (Victoria University Wellington), Yi Mei (Victoria University Wellington), Mengjie Zhang (Victoria University Wellington)
关键词：交通信号控制、遗传编程
RWA Track
赛道：Real World Applications
题目：Quality Diversity Approaches for Time-Use Optimisation to Improve Health Outcomes
作者：Adel Nikfarjam (The University of Adelaide), Ty Stanford (The University of South Australia), Aneta Neumann (The University of Adelaide), Dorothea Dumuid (The University of South Australia), Frank Neumann (The University of Adelaide)
关键词：幸福感分析、质量多样性优化
Student Workshop Award
题目：Evolving Quantum Logic Gate Circuits in Qiskit
作者：Thomas Robert Newbold (University of Exeter), Alberto Moraglio (University of Exeter)
关键词：量子逻辑门电路、遗传编程
简单总结
在今年的十篇最佳论文中，五篇是演化计算求解特定应用问题，三篇是演化算法性能的提升和改进、两篇是运行复杂度分析。
从每年录取的GECCO论文来看，应用问题层出不穷，这些文章有助于我们了解演化计算在不同科学领域的实际作用，这两年随着传统科学和人工智能技术的不断融合，EC for Science有可能迎来高潮。
GECCO上最火热的算法改进文章包括演化策略、质量多样性优化、蚁群优化、多目标优化、增强拓扑神经进化等。这些方法经历多年发展已形成自己独特的方法论，尤其是演化策略和质量多样性优化，包括理论解释、算法设计原则、基础软件平台、benchmark、并行化加速等等。这些方法本身依旧面临实用性挑战，还需要进一步研究和完善。
运行复杂度分析是近些年来理论演化计算最火热的话题之一，从遗传算法到蚁群优化、从数值优化到组合优化、从单目标优化到多目标优化。各种演化算法在理论benchmark上的运行复杂度分析为演化算法的推广提供坚实的理论基础。
