Evolutionary algorithms (EAs) have been a popular optimization tool for two decades in academia. However, the industry applications of EAs to real-world optimization problems are infrequent, due to the strong assumption that objective function evaluations are straightforward and easily accessed. In fact, such objective functions may not exist, instead computationally expensive numerical simulations or costly physical experiments must be performed for evaluations. In more extreme cases, only historical data are available for performing optimization and no new data can be generated during optimization. Such problems driven by only historical data are formulated as offline data-driven optimization problems, which poses challenges to conventional EAs. Although surrogate models that approximate objective functions can be used as function evaluations in EAs to guide the search, the search accuracy is limited by the offline data and cannot be validated or enhanced by sampling new data. The research on offline data-driven evolutionary optimization has not received sufficient attention, although techniques for solving such problems are very practical. One main reason is the lack of benchmark problems that can closely reflect real-world challenges, which leads to a big gap between academia and industries.
In this competition, there are two tracks: continuous single- and multi-objective optimization. We carefully select three benchmark problems for each track from an interesting real-world application: parameter optimization of flocking model for a swarm of robots. The optimized flocking mode is used to control a group of autonomous drones in three typical scenarios: unconfined and obstacle-free environment, confined and obstacle-free environment, tunnel environment with an obstacle. The objective functions of those problems cannot be calculated analytically, but can be calculated by calling a simulation to provide true black-box evaluations for offline data sampling and the final comparison. Participants are required to find the optimal solutions of each benchmark problem based on the provided offline data. To assess the performance of compared algorithms, we will evaluate the found optimal solutions using the same simulation which generates the offline data.
Track 1: continuous single-objective optimization:
DDSOP1: This problem is a collective behavior performance maximization of a group of autonomous drones flying in unconfined and obstacle-free environment, which has 4 decision variables.
DDSOP2: This problem is a collective behavior performance maximization of a group of autonomous drones flying in confined and obstacle-free environment, which has 4 decision variables.
DDSOP3: This problem is a collective behavior performance maximization of a group of autonomous drones flying in tunnel environment with an obstacle, which has 4 decision variables.
Track 2: continuous multi-objective optimization:
DDMOP1: This problem is a collective behavior performance maximization of a group of autonomous drones flying in unconfined and obstacle-free environment, which has 4 decision variables and 3 objectives.
DDMOP2: This problem is a collective behavior performance maximization of a group of autonomous drones flying in confined and obstacle-free environment, which has 4 decision variables and 3 objectives.
DDMOP3: This problem is a collective behavior performance maximization of a group of autonomous drones flying in tunnel environment with an obstacle, which has 4 decision variables and 3 objectives
For both tracks, the boundary of decision variable 1,2,4 is the range between 0.0001 and 100, the boundary of decision variable 3 is the range between 0.0001 and 300.
Download links: https://github.com/HandingWang/DDEO-CEC2021
Important Dates:
For participants planning to submit a paper to the 2021 IEEE Congress on Evolutionary Computation:
Paper submission: 31 Jan 2021
Notification to authors: 22 Mar 2021
Final submission: 7 Apr 2021
Note: You are encouraged to submit your paper to the Special Session on Data-Driven Optimization of Computationally Expensive Problems
For other participants (only result entry but without a paper):
Results submission deadline: 1 May 2021
Note: Please send your results directly to Dr Handing Wang (hdwang@xidian.edu.cn)
Organizers::
Handing Wang, School of Artificial Intelligence, Xidian University, China
Yanan Li, School of Marine Science and Technology, Northwestern Polytechnical University, China
Xingguang Peng,School of Marine Science and Technology, Northwestern Polytechnical University, China
Yaochu Jin, Department of Computer Science, University of Surrey, UK
Competition entries::
Track 1: continuous single-objective optimization:
SAPSO: Surrogate-Assisted Particle Swarm Optimization, Jie Liu, Weixi Chen, Huachao Dong, Peng Wang, Northwestern Polytechnical University
MS-DDEA: offline data-driven evolutionary algorithm based on model selection, Huixiang Zhen, Wenyin Gong, Ling Wang, China University of Geosciences
DAK-Opt: DE & ACO hybrid optimization algorithm based on Kriging Jia Xuyi, Li Chunna, Northwestern Polytechnical University
DDEA-SE: Data-Driven Evolutionary Optimization Using Selective Surrogate Ensembles, Tuo Zhang, Handing Wang Xidian University
KSOOA: Kriging-based Single objective optimization Algorithm Liang Fan, Handing Wang, Xidian University
DDO-RBFNC: Data-driven optimization based on radial basis functions with neighborhood clustering, Junpeng Su, Han Huang South China University of Technology
C-DDEA: Comparison-based offline data-driven EA with local?surrogate Haogan Huang, South China University of Technology
DDEA-XGBS: Data-Driven Evolutionary Algorithm with xgboost surrogate, Yongcun Liu, Handing Wang Xidian University
DDEA-SG: Offine Data-Driven Evolutionary Algorithm based on Stacked Generalization, Xijun Huang, Ming Chen, ShenTian Li, ZhengPing Liang Shenzhen University
HVEA: Hypervolume evolutionary algorithm Lei Han, Handing Wang, Xidian University
SurroOpt: surrogate-based optimization method xu chenzhou, dang haodong, liu mingqi, han zhong hua, Northwestern Polytechnical University
BO-PP(X): Bayesian Optimization using Pseudo-Points (Xgboost) Zhihong Qi, Yufei Xiang, Yawen Zhou, Ke Xue, Chao Qian, Nanjing University
Track 2: continuous multi-objective optimization:
SACEA: Surrogate-Assisted Muti-objective Co-Evolutionary Algorithm, Jiangtao Shen, Jinglu Li, Wenxin Wang, Huachao Dong, Peng Wang Northwestern Polytechnical University
IBEA-MS: Performance Indicator based Adaptive Model Selection for Offline Data-Driven Multi-Objective Evolutionary Optimization, Zhening Liu, Handing Wang Xidian University
RF-2REA: A Random Forest-Assisted Two Round Selection Evolutionary Algorithm for Expensive Multi-Objective Optimization, Ming Chen, Xijun Huang, Shentian Li, Zhengping Liang, Shenzhen University
DDEA-SG: Offine Data-Driven Evolutionary Algorithm based on Stacked Generalization, Xijun Huang, Ming Chen, ShenTian Li, ZhengPing Liang Shenzhen University
KTA2: Kriging-assisted Two_Arch2, Zhenshou Song, Junfeng Tang, Handing Wang Xidian University
PBNSGAIII: Kriging-assisted NSGAIII, Zhenshou Song, Junfeng Tang, Handing Wang Xidian University
NSGA-II(X): Non-dominated Sorting Genetic Algorithm II (Xgboost), Zhihong Qi, Yufei Xiang, Yawen Zhou, Ke Xue, Chao Qian Nanjing University
ODDMOEA-BC: Offline data-driven multi-objective optimization based on boundary solution culling, Lianbo Yang, Yang Liu, Chunna Li, Northwestern Polytechnical University
All the results can be found at: https://github.com/HandingWang/DDEO-CEC2021
Winer algorithms for Track 1:
Winner: DDO-RBFNC
First runner-up: DDEA-SG
Second runner-up: SurroOpt
Winer algorithms for Track 2:
Winner: ODDMOEA-BC
First runner-up: RF-2REA
Second runner-up: SACEA