计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (9): 299-309.doi: 10.11896/jsjkx.230600074

• 人工智能 • 上一篇    下一篇

基于时空间联合去噪的改进差分进化算法

王彬1, 张鑫雨1, 金海燕1,2   

  1. 1 西安理工大学计算机科学与工程学院 西安 710048
    2 西安理工大学陕西省网络计算与安全技术重点实验室 西安 710048
  • 收稿日期:2023-06-08 修回日期:2023-11-19 出版日期:2024-09-15 发布日期:2024-09-10
  • 通讯作者: 金海燕(jinhaiyan@xaut.edu.cn)
  • 基金资助:
    国家自然科学基金(62272383,62372369)

Improved Differential Evolution Algorithm Based on Time-Space Joint Denoising

WANG Bin1, ZHANG Xinyu1, JIN Haiyan1,2   

  1. 1 School of Computer Science and Engineering,Xi'an University of Technology,Xi'an 710048,China
    2 Shaanxi Key Laboratory for Network Computing and Security Technology,Xi'an University of Technology,Xi'an 710048,China
  • Received:2023-06-08 Revised:2023-11-19 Online:2024-09-15 Published:2024-09-10
  • About author:WANG Bin,born in 1971,associate professor,master supervisor,is a senior member of CCF(No.14802S).His main research interests include evolutionary computation and artificial intelligence.
    JIN Haiyan,born in 1976,professor,Ph.D supervisor,is an outstanding member of CCF(No.74449M).Her main research interests include artificial intelligence theory and its applications and so on.
  • Supported by:
    National Natural Science Foundation of China(62272383,62372369).

PDF (PC)

328

摘要： 在工程问题的优化求解过程中,对个体的适应度评价可能会受到环境噪声的干扰,进而影响对种群进行合理的优胜劣汰操作,造成算法性能下降。为了对抗噪声环境的影响,提出了一种基于时空间联合去噪的改进差分进化算法(SEDADE)。根据适应度排名将种群划分成两个子种群,对评价较差个体组成的子种群用分布估计算法(EDA)进化,采用高斯分布建模解空间,利用解空间中多个个体噪声的随机性抵消噪声影响;对评价较好个体组成的子种群用差分进化算法(DE)进化,并且引入基于时间的停滞重采样机制去噪,提高收敛精度。对时空间混合进化得到的两个子种群进行基于概率选择的EDA信息利用操作,利用EDA搜索得到的全局信息引导DE的搜索方向,避免陷入局部最优。在实验中使用了被零均值高斯噪声干扰的基准函数,可以发现SEDADE相比其他算法更具有竞争性,此外通过消融实验验证了所提算法包含的3个机制的有效性和合理性。

关键词: 差分进化, 分布估计, 噪声, 重采样, 混合进化, 信息利用

Abstract: In the optimization process of solving engineering problems,the evaluation of individual fitness may be affected by environmental noise,so as to affect the reasonable survival of the fittest operation on the population,and result in a decline in algorithm performance.In order to combat the impact of noise environment,an improved differential evolution algorithm(SEDADE) based on joint temporal and spatial denoising is proposed.The population is divided into two subpopulations according to fitness ranking,and the subpopulations composed of poorly evaluated individuals are evolved using a distribution estimation algorithm(EDA).Gaussian distribution is used to model the solution space,using the randomness of multiple individual noises in the solution space to offset the noise impact.Differential evolution algorithm(DE) is used to evolve subpopulations with better evaluated individual composition,and a time-based stagnation resampling mechanism is introduced to denoise to improve convergence accuracy.The EDA information utilization operation based on probability selection is performed on the two subpopulations derived from time-space mixed evolution,and the global information obtained from EDA search is used to guide the search direction of DE to avoid falling into local optimization.In the experiment,a benchmark function interfered by zero mean Gaussian noise is used,and it is found that SEDADE is competitive with other algorithms.In addition,the effectiveness and rationality of the proposed mechanism are verified through ablation experiments.

Key words: Differential evolution, Distribution estimation, Noise, Resampling, Hybrid evolution, Information utilization

中图分类号: 

  • TP301
[1]VIKHAR P A.Evolutionary algorithms:A critical review and its future prospects[C]//International Conference on Global Trends in Signal Processing,Information Computing and Communication(ICGTSPICC).IEEE,2016:261-265.
[2]TAU Y H S.Optimization of power delivery design and metho-dologies[C]//International Conference on Electronic Materials and Packaging.IEEE,2006:1-3.
[3]LI Z,ZHANG S,CAI X,et al.Noisy Optimization by Evolution Strategies With Online Population Size Learning[J].IEEE Transactions on Systems,Man,and Cybernetics:Systems,2022,52(9):5816-5828.
[4]CHIU S Y,LIN C N,LIU J,et al.Differential evolution forstrongly noisy optimization:Use 1.01 n resamplings at iteration n and reach the 1/2 slope[C]//IEEE Congress on Evolutionary Computation(CEC).IEEE,2015:338-345.
[5]GUO D,NIE Z,YAN L.The application of noise-tolerant ZD design formula to robots' kinematic control via time-varying nonlinear equations solving[J].IEEE Transactions on Systems,Man,and Cybernetics:Systems,2018,48(12):2188-2197.
[6]JING B B,GUO J,WANG L Q,et al.Image double blind denoi-sing algorithm combining with denoising convolutional neural network and conditional generative adversarial net[J].Journal of Computer Applications,2021,41(6):1767-1774.
[7]RAKSHIT P,KONAR A,DAS S,et al.Uncertainty Management in Differential Evolution Induced Multiobjective Optimization in Presence of Measurement Noise[J].IEEE Transactions on Systems Man & Cybernetics Systems,2017,44(7):922-937.
[8]MERELO J J,LIBERATORE F,ARES A F,et al.There isNoisy Lunch:A Study of Noise in Evolutionary Optimization Problems[C]//International Joint Conference on Computational Intelligence.IEEE,2016.
[9]JIN Y,BRANKE J.Evolutionary optimization in uncertain environments-a survey[J].IEEE Transactions on Evolutionary Computation,2005,9(3):303-317.
[10]CAPONIO A,NERI F.Differential Evolution with Noise Analyzer[C]//Workshops on Applications of Evolutionary Computation.Berlin,Heidelberg: Springer Berlin Heidelberg,2009:715-724.
[11]MININNO E,NERI F.A memetic Differential Evolution ap-proach in noisy optimization[J].Memetic Computing,2010,2(2):111-135.
[12]DI PIETRO A,WHILE L,BARONE L.Applying evolutionaryalgorithms to problems with noisy,time-consuming fitness functions[C]//Proceedings of the 2004 Congress on Evolutionary Computation(IEEE Cat.No.04TH8753).IEEE,2004:1254-1261.
[13]MARKON S,ARNOLD D V,BACK T,et al.Thresholding－aselection operator for noisy ES[C]//Proceedings of the 2001 Congress on Evolutionary Computation.IEEE,2001:465-472.
[14]DAS S,KONAR A,CHAKRABORTY U K.Improved differential evolution algorithms for handling noisy optimization pro-blems[C]//2005 IEEE Congress on Evolutionary Computation.IEEE,2005:1691-1698.
[15]DAS S,KONAR A.An Improved Differential Evolution Scheme for Noisy Optimization Problems[C]//International Conference on Pattern Recognition & Machine Intelligence.Springer-Verlag,2005.
[16]BREST J,GREINER S,BOSKOVIC B,et al.Self-adapting con-trol parameters in differential evolution:A comparative study on numerical benchmark problems[J].IEEE Transactions on Evolutionary Computation,2006,10(6):646-657.
[17]RAHNAMAYAN S,TIZHOOSH H R,SALAMA M M A.Opposition-based differential evolution for optimization of noisy problems[C]//2006 IEEE International Conference on Evolutionary Computation.IEEE,2006:1865-1872.
[18]GHOSH A,DAS S,PANIGRAHI B K,et al.A noise resilient differential evolution with improved parameter and strategy control[C]//IEEE congress on evolutionary computation(CEC).IEEE,2017:2590-2597.
[19]ZHANG J,ZHU X,WANG Y,et al.Dual-Environmental Particle Swarm Optimizer in Noisy and Noise-Free Environments [J].IEEE Transactions on Cybernetics,2019,49(6):2011-2021.
[20]ZHAO Q,GAO Y.A new Algorithm based on the Gbest of Particle Swarm Optimization algorithm to improve Estimation of Distribution Algorithm[C]//2018 International Conference on Smart Computing and Electronic Enterprise(ICSCEE).IEEE,2018:1-5.
[21]QIU L Q,LIANG Y Q,FAN J C.A Dynamic Fusion Parallel Hybrid Evolutionary Algorithm EDAs/PSO [J].Computer Application and Software,2014,31(6):271-274.
[22]SUN J,ZHANG Q,TSANG E P K.DE/EDA:A new evolu-tionary algorithm for global optimization[J].Information Sciences,2005,169(3／4):249-262.
[23]STORN R,PRICE K.Differential Evolution:A Simple and Efficient Adaptive Scheme for Global Optimization Over Continuous Spaces[J].Journal of Global Optimization,1995,23(1):1-12.
[24]DAS S,SUGANTHAN P N.Differential evolution:A survey of the state-of-the-art[J].IEEE Transactions on Evolutionary Computation,2010,15(1):4-31.
[25]DAS S,MULLICK S S,SUGANTHAN P N.Recent advances in differential evolution-an updated survey[J].Swarm and Evolutionary Computation,2016,27:1-30.
[26]MUHLENBEIN H.The equation for response to selection and its use for prediction[J].Evolutionary Computation,1997,5(3):303-346.
[27]OCHOA A.EBBA－Evolutionary best basis algorithm[C]//Proceedings of the Second International Symposium on Adaptive Systems(ISAS 99),Havana,Cuba.1999:93-98.
[28]LARRANAGA P,LOZANO J A.Estimation of distribution algorithms:A new tool for evolutionary computation[M].Boston:Kluwer Press,2002.
[29]ZHANG B T.A Bayesian Framework for Evolutionary Computation[C]//Proceedings of the 1999 Congress on Evolutionary Conputation(CEC99).1999:722-728.
[30]BALUJA S.Population-based incremental learning.a method for integrating genetic search based function optimization and competitive learning[R].Carnegie-Mellon Univ Pittsburgh Pa Dept of Computer Science,1994.
[31]RAKSHIT P,KONAR A,DAS S.Noisy evolutionary optimization algorithms－a comprehensive survey[J].Swarm and Evolutionary Computation,2017,33:18-45.
[32]KRINK T,FILIPIC B,FOGEL G B.Noisy optimization pro-blems－a particular challenge for differential evolution?[C]//Proceedings of the 2004 Congress on Evolutionary Computation(IEEE Cat.No.04TH8753).IEEE,2004,1:332-339.
[33]GHOSH A,DAS S,MALLIPEDDI R,et al.A modified differential evolution with distance-based selection for continuous optimization in presence of noise[J].IEEE Access,2017,5:26944-26964.
[34]GUO X,YANG Q,ZHENG H,et al.Optimization of power distribution for electrothermal anti-icing systems by differential evolution algorithm[J].Applied Thermal Engineering,2023,221:119875.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发