[1]李 晰,李 帅,冯艳红,等.基于联合分布适配的单向迁移差分进化算法[J].郑州大学学报(工学版),2023,44(05):24-31.[doi:10.13705/j.issn.1671-6833.2023.05.005]
　LI Xi,LI Shuai,FENG Yanhong,et al.Unidirectional Transfer Differential Evolution Algorithm Based on Joint Distribution Adaptation[J].Journal of Zhengzhou University (Engineering Science),2023,44(05):24-31.[doi:10.13705/j.issn.1671-6833.2023.05.005]
点击复制

基于联合分布适配的单向迁移差分进化算法()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
44
期数:
2023年05期
页码:
24-31
栏目:
出版日期:
2023-08-20

文章信息/Info

Title:
Unidirectional Transfer Differential Evolution Algorithm Based on Joint Distribution Adaptation
作者:
李 晰1; 2 ;  李 帅1; 2 ;  冯艳红1 ;  李明亮1
1. 河北地质大学 信息工程学院,河北 石家庄 050031;2. 河北地质大学 人工智能与机器学习研究室,河北 石家庄 050031
Author(s):
LI Xi ;  LI Shuai;  FENG Yanhong;  LI Mingliang
关键词:
优化算法;  迁移学习;  联合分布适配;  单向迁移;  差分进化算法
Keywords:
ptimization algorithms;  transfer learning;  joint distribution adaptation;  unidirectional transfer;  differential evolutionary
分类号:
O234；TP301
DOI:
10.13705/j.issn.1671-6833.2023.05.005
文献标志码:
A
摘要:
传统的差分进化算法求解优化问题一般从零知识开始,独立搜索,没有利用已求解过的相似问题信息,针 对这一问题,在传统的差分进化算法中引入迁移学习技术。 首先,利用存在相关性的源问题的优化种群和目标问 题的当前种群抽取关键信息,通过联合分布适配的方法映射到高维希尔伯特空间。 其次,用映射后得到的矩阵构 建新种群,代替目标问题的种群,完成后续进化任务。 实现了 2 种迁移模式:在目标问题求解初始化时,将源问题 的有效信息进行迁移,引导算法搜索方向;目标问题求解迭代一定的次数后,再利用迁移的有效信息,加快种群收 敛速度。 最后,采用 9 组多任务测试函数对算法进行了测试,与无迁移的差分进化算法以及直接迁移种群的无适 配技术的差分进化算法进行对比。 结果表明:在求解质量方面,所提算法有 7 组优于传统的无迁移差分进化算法; 在求解速度方面,所提算法有 7 组比传统差分进化算法收敛速度更快;基于迁移学习的差分进化算法对提高目标 优化问题的求解精度和收敛速度是有效的。
Abstract:
In traditional differential evolution algorithm solves optimization problems generally starting from zeroknowledge and searching independently without using the information of similar problems that have been solved. In this paper, the transfer learning technique was introduced into DE. Firstly, the extracted key information from the optimized population of the source problem and the current population of the target problem was mapped to the highdimensional Hilbert space by the joint distribution adaptation method. Then a new population was constructed from the mapped matrix to replace the population of the target problem. Lastly, the subsequent evolution was completed. Two transfer modes were implemented: transferring the effective information of the source problem to guide the search direction of the algorithm during the initialization; utilizing the transferred effective information after a certain number of iterations to accelerate the population convergence. The proposed algorithm was tested with nine sets of multi-task test functions, and compared with the DE without transfer and transferred DE without adaptation technique. The results showed that, the proposed algorithm outperforms the traditional DE on seven sets functions in term of solution quality. Moreover, the proposed algorithm converged faster than DE. Therefore, the transfer learning-based differential evolution algorithm was effective in improving the solution accuracy and convergence speed of the objective optimization problem.

参考文献/References:

[1] DAS S, SUGANTHAN P N. Differential evolution: a survey of the state-of-the-art[ J] . IEEE Transactions on Evolutionary Computation, 2011, 15(1) : 4-31.

 [2] 程适, 王锐, 伍国华, 等. 群体智能优化算法[ J] . 郑 州大学学报(工学版) , 2018,39(6) : 1-2. 
CHENG S,WANG R,WU G H,et al. Swarm intelligence optimization algorithm[ J] . Journal of Zhengzhou University (Engineering Science) , 2018,39(6) : 1-2.
 [3] 刘倩, 冯艳红, 陈嶷瑛. 基于混沌初始化和高斯变异 的飞蛾火焰优化算法[ J] . 郑州大学学报( 工学版) , 2021, 42(3) : 53-58. 
LIU Q, FENG Y H, CHEN Y Y. Moth-flame optimization algorithm based on chaotic initialization and Gaussian mutation[ J] . Journal of Zhengzhou University (Engineering Science) , 2021, 42(3) : 53-58. 
[4] 汪慎文, 张佳星, 褚晓凯, 等. 两阶段搜索的多模态 多目标差分进化算法[ J] . 郑州大学学报( 工学版) , 2021, 42(1) : 9-14, 110. 
WANG S W, ZHANG J X, CHU X K, et al. Multimodal multi-objective differential evolution algorithm based on two-stage search [ J ] . Journal of Zhengzhou University (Engineering Science) , 2021, 42(1) : 9-14, 110.
 [5] TAN Z P, LI K S, WANG Y. Differential evolution with adaptive mutation strategy based on fitness landscape analysis[J]. Information Sciences, 2021, 549: 142-163. 
[6] PAN S J, YANG Q. A survey on transfer learning [ J] . IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10) : 1345-1359. 
[7] XU H P, YANG L Z, LONG X Y. Underwater sonar image classification with small samples based on parameterbased transfer learning and deep learning [ C ] ∥2022 Global Conference on Robotics, Artificial Intelligence and Information Technology ( GCRAIT) . Piscataway: IEEE, 2022: 304-307.
 [8] CHIBA S, SASAOKA H. Basic study for transfer learning for autonomous driving in car race of model car[C]∥2021 6th International Conference on Business and Industrial Research ( ICBIR ) . Piscataway: IEEE, 2021: 138 -141.
 [9] SINGH N K, KUMAR A. Image based recommender system using transfer learning [ C] ∥2022 2nd International Conference on Emerging Frontiers in Electrical and Electronic Technologies ( ICEFEET ) . Piscataway: IEEE, 2022: 1-4. 
[10] TAN K C, FENG L, JIANG M. Evolutionary transfer optimization-a new frontier in evolutionary computation re- search[ J] . IEEE Computational Intelligence Magazine, 2021, 16(1) : 22-33.
 [11] LIM R, GUPTA A, ONG Y S, et al. Non-linear domain adaptation in transfer evolutionary optimization[ J] . Cognitive Computation, 2021, 13(2) : 290-307.
 [12] JIANG M, HUANG Z Q, QIU L M, et al. Transfer learning-based dynamic multiobjective optimization algorithms[ J] . IEEE Transactions on Evolutionary Computation, 2018, 22(4) : 501-514. 
[13] JIANG M, QIU L M, HUANG Z Q, et al. Dynamic multi-objective estimation of distribution algorithm based on domain adaptation and nonparametric estimation [ J] . Information Sciences, 2018, 435: 203-223.
 [14] 张勇, 杨康, 郝国生, 等. 基于相似历史信息迁移学习的 进化优化框架[J]. 自动化学报, 2021, 47(3): 652-665.
 ZHANG Y, YANG K, HAO G S, et al. Evolutionary optimization framework based on transfer learning of similar historical information[ J] . Acta Automatica Sinica, 2021, 47(3) : 652-665. 
[15] 刘璐, 蒋艳. 基于迁移学习的 NSGAⅡ算法[ J] . 软件 导刊, 2021, 20(3) : 134-138. 
LIU L, JIANG Y. NSGAⅡ algorithm based on transfer learning[ J] . Software Guide, 2021, 20(3) : 134-138. 
[16] 尚青霞, 周磊, 冯亮. 基于降噪自动编码器的多任务优 化算法[J]. 大连理工大学学报, 2019, 59(4): 417-426. 
SHANG Q X, ZHOU L, FENG L. Multi-task optimization algorithm based on denoising auto-encoder[ J] . Journal of Dalian University of Technology, 2019, 59 ( 4) : 417-426. 
[17] FENG L, ZHOU L, ZHONG J H, et al. Evolutionary multitasking via explicit autoencoding[ J] . IEEE Transactions on Cybernetics, 2019, 49(9) : 3457-3470.
 [18] PAN S J, TSANG I W, KWOK J T, et al. Domain adaptation via transfer component analysis[ J] . IEEE Transactions on Neural Networks, 2011, 22(2) : 199-210.
 [19] GONG B Q, SHI Y, SHA F, et al. Geodesic flow kernel for unsupervised domain adaptation [ C] ∥IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2012: 2066-2073.
 [20] LONG M S, WANG J M, DING G G, et al. Transfer feature learning with joint distribution adaptation[ C]∥2013 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2014: 2200-2207. 
[21] DA B S, ONG Y S, FENG L, et al. Evolutionary multitasking for single-objective continuous optimization: benchmark problems, performance metric and baseline results[EB / OL] . (2017-06-12) [ 2022-10-26] . https:∥ doi. org / 10. 48550 / arXiv. 1706. 03470.

更新日期/Last Update: 2023-09-04