用于回归演化特征构建的几何语义宏交叉算子

IF 1.7 3区 计算机科学 Q3 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE
Hengzhe Zhang, Qi Chen, Bing Xue, Wolfgang Banzhaf, Mengjie Zhang
{"title":"用于回归演化特征构建的几何语义宏交叉算子","authors":"Hengzhe Zhang, Qi Chen, Bing Xue, Wolfgang Banzhaf, Mengjie Zhang","doi":"10.1007/s10710-023-09465-z","DOIUrl":null,"url":null,"abstract":"<p>Evolutionary feature construction has been successfully applied to various scenarios. In particular, multi-tree genetic programming-based feature construction methods have demonstrated promising results. However, existing crossover operators in multi-tree genetic programming mainly focus on exchanging genetic materials between two trees, neglecting the interaction between multi-trees within an individual. To increase search effectiveness, we take inspiration from the geometric semantic crossover operator used in single-tree genetic programming and propose a macro geometric semantic crossover operator for multi-tree genetic programming. This operator is designed for feature construction, with the goal of generating offspring containing informative and complementary features. Our experiments on 98 regression datasets show that the proposed geometric semantic macro-crossover operator significantly improves the predictive performance of the constructed features. Moreover, experiments conducted on a state-of-the-art regression benchmark demonstrate that multi-tree genetic programming with the geometric semantic macro-crossover operator can significantly outperform all 22 machine learning algorithms on the benchmark.</p>","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"17 1","pages":""},"PeriodicalIF":1.7000,"publicationDate":"2023-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A geometric semantic macro-crossover operator for evolutionary feature construction in regression\",\"authors\":\"Hengzhe Zhang, Qi Chen, Bing Xue, Wolfgang Banzhaf, Mengjie Zhang\",\"doi\":\"10.1007/s10710-023-09465-z\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>Evolutionary feature construction has been successfully applied to various scenarios. In particular, multi-tree genetic programming-based feature construction methods have demonstrated promising results. However, existing crossover operators in multi-tree genetic programming mainly focus on exchanging genetic materials between two trees, neglecting the interaction between multi-trees within an individual. To increase search effectiveness, we take inspiration from the geometric semantic crossover operator used in single-tree genetic programming and propose a macro geometric semantic crossover operator for multi-tree genetic programming. This operator is designed for feature construction, with the goal of generating offspring containing informative and complementary features. Our experiments on 98 regression datasets show that the proposed geometric semantic macro-crossover operator significantly improves the predictive performance of the constructed features. Moreover, experiments conducted on a state-of-the-art regression benchmark demonstrate that multi-tree genetic programming with the geometric semantic macro-crossover operator can significantly outperform all 22 machine learning algorithms on the benchmark.</p>\",\"PeriodicalId\":50424,\"journal\":{\"name\":\"Genetic Programming and Evolvable Machines\",\"volume\":\"17 1\",\"pages\":\"\"},\"PeriodicalIF\":1.7000,\"publicationDate\":\"2023-12-08\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Genetic Programming and Evolvable Machines\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://doi.org/10.1007/s10710-023-09465-z\",\"RegionNum\":3,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Genetic Programming and Evolvable Machines","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1007/s10710-023-09465-z","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0

摘要

进化特征构建已成功应用于各种场景。其中,基于多树遗传编程的特征构建方法取得了可喜的成果。然而,现有的多树遗传编程中的交叉算子主要集中于两棵树之间的遗传物质交换,而忽略了个体内部多树之间的相互作用。为了提高搜索效率,我们从单树遗传编程中使用的几何语义交叉算子中汲取灵感,提出了一种适用于多树遗传编程的宏几何语义交叉算子。该算子专为特征构建而设计,目标是生成包含信息丰富且互补特征的后代。我们在 98 个回归数据集上的实验表明,所提出的几何语义宏交叉算子显著提高了所构建特征的预测性能。此外,在最先进的回归基准上进行的实验表明,使用几何语义宏交叉算子的多树遗传编程在该基准上明显优于所有 22 种机器学习算法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

A geometric semantic macro-crossover operator for evolutionary feature construction in regression

A geometric semantic macro-crossover operator for evolutionary feature construction in regression

Evolutionary feature construction has been successfully applied to various scenarios. In particular, multi-tree genetic programming-based feature construction methods have demonstrated promising results. However, existing crossover operators in multi-tree genetic programming mainly focus on exchanging genetic materials between two trees, neglecting the interaction between multi-trees within an individual. To increase search effectiveness, we take inspiration from the geometric semantic crossover operator used in single-tree genetic programming and propose a macro geometric semantic crossover operator for multi-tree genetic programming. This operator is designed for feature construction, with the goal of generating offspring containing informative and complementary features. Our experiments on 98 regression datasets show that the proposed geometric semantic macro-crossover operator significantly improves the predictive performance of the constructed features. Moreover, experiments conducted on a state-of-the-art regression benchmark demonstrate that multi-tree genetic programming with the geometric semantic macro-crossover operator can significantly outperform all 22 machine learning algorithms on the benchmark.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Genetic Programming and Evolvable Machines
Genetic Programming and Evolvable Machines 工程技术-计算机:理论方法
CiteScore
5.90
自引率
3.80%
发文量
19
审稿时长
6 months
期刊介绍: A unique source reporting on methods for artificial evolution of programs and machines... Reports innovative and significant progress in automatic evolution of software and hardware. Features both theoretical and application papers. Covers hardware implementations, artificial life, molecular computing and emergent computation techniques. Examines such related topics as evolutionary algorithms with variable-size genomes, alternate methods of program induction, approaches to engineering systems development based on embryology, morphogenesis or other techniques inspired by adaptive natural systems.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信