Proceedings of the Companion Conference on Genetic and Evolutionary Computation最新文献_第10页

Phased Genetic Programming for Application to the Traveling Salesman Problem 阶段遗传规划在旅行商问题中的应用

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3590673

D. Chitty, E. Keedwell

{"title":"Phased Genetic Programming for Application to the Traveling Salesman Problem","authors":"D. Chitty, E. Keedwell","doi":"10.1145/3583133.3590673","DOIUrl":"https://doi.org/10.1145/3583133.3590673","url":null,"abstract":"The Traveling Salesman Problem (TSP) is a difficult permutation-based optimisation problem typically solved using heuristics or meta-heuristics which search the solution problem space. An alternative is to find sets of manipulations to a solution which lead to optimality. Hyper-heuristics search this space applying heuristics sequentially, similar to a program. Genetic Programming (GP) evolves programs typically for classification or regression problems. This paper hypothesizes that GP can be used to evolve heuristic programs to directly solve the TSP. However, evolving a full program to solve the TSP is likely difficult due to required length and complexity. Consequently, a phased GP method is proposed whereby after a phase of generations the best program is saved and executed. The subsequent generation phase restarts operating on this saved program output. A full program is evolved piecemeal. Experiments demonstrate that whilst pure GP cannot solve TSP instances when using simple operators, Phased-GP can obtain solutions within 4% of optimal for TSPs of several hundred cities. Moreover, Phased-GP operates up to nine times faster than pure GP.","PeriodicalId":422029,"journal":{"name":"Proceedings of the Companion Conference on Genetic and Evolutionary Computation","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127429422","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Co-evolving Recurrent Neural Networks and their Hyperparameters with Simplex Hyperparameter Optimization 基于单纯形超参数优化的协同进化递归神经网络及其超参数

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3596407

A. Kini, S. Yadav, Aditya Shankar Thakur, A. Awari, Zimeng Lyu, Travis J. Desell

{"title":"Co-evolving Recurrent Neural Networks and their Hyperparameters with Simplex Hyperparameter Optimization","authors":"A. Kini, S. Yadav, Aditya Shankar Thakur, A. Awari, Zimeng Lyu, Travis J. Desell","doi":"10.1145/3583133.3596407","DOIUrl":"https://doi.org/10.1145/3583133.3596407","url":null,"abstract":"Designing machine learning models involves determining not only the network architecture, but also non-architectural elements such as training hyperparameters. Further confounding this problem, different architectures and datasets will perform more optimally with different hyperparameters. This problem is exacerbated for neuroevolution (NE) and neural architecture search (NAS) algorithms, which can generate and train architectures with a wide variety of architectures in order to find optimal architectures. In such algorithms, if hyperparameters are fixed, then suboptimal architectures can be found as they will be biased towards the fixed parameters. This paper evaluates the use of the simplex hyperparameter optimization (SHO) method, which allows co-evolution of hyperparameters over the course of a NE algorithm, allowing the NE algorithm to simultaneously optimize both network architectures and hyperparameters. SHO has been previously shown to be able to optimize hyperparameters for convolutional neural networks using traditional stochastic gradient descent with Nesterov momentum, and this work extends on this to evaluate SHO for evolving recurrent neural networks with additional modern weight optimizers such as RMSProp and Adam. Results show that incorporating SHO into the neuroevolution process not only enables finding better performing architectures but also faster convergence to optimal architectures across all datasets and optimization methods tested.","PeriodicalId":422029,"journal":{"name":"Proceedings of the Companion Conference on Genetic and Evolutionary Computation","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127473030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evolving Nonlinear Multigrid Methods With Grammar-Guided Genetic Programming 基于语法导向遗传规划的非线性多网格演化方法

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3590734

Dinesh Parthasarathy, J. Schmitt, H. Köstler

引用次数: 0

Extending Tree-Based Automated Machine Learning to Biomedical Image and Text Data Using Custom Feature Extractors 使用自定义特征提取器将基于树的自动机器学习扩展到生物医学图像和文本数据

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3590584

Rachit Kumar, Joseph D. Romano, M. Ritchie, Jason Moore

引用次数: 0

Repetitive Processes and Their Surrogate-Model Congruent Encoding for Evolutionary Algorithms - A Theoretic Proposal 进化算法的重复过程及其代理模型全同编码——一个理论建议

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3596389

Christina Plump, Bernhard Berger, R. Drechsler

引用次数: 0

Fast Collective Decision-Making without Prior Knowledge 没有先验知识的快速集体决策

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3590623

Nicolas Cambier, E. Ferrante

引用次数: 0

Benchmarking and analyzing iterative optimization heuristics with IOHprofiler 基于IOHprofiler的迭代优化启发式测试与分析

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3595057

Carola Doerr, Hongya Wang, Diederick Vermetten, Thomas Bäck, Jacob De Nobel, Furong Ye

引用次数: 0

Large-Scale Optimization and Learning 大规模优化与学习

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3595037

M. Omidvar, Yuan Su, Xiaodong Li

引用次数: 0

Pareto Front Upconvert on Multi-objective Building Facility Control Optimization 多目标楼宇设施控制优化的Pareto前上转换

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3596339

Naru Okumura, Tomoaki Takagi, Yoshihiro Ohta, Hiroyuki Sato

{"title":"Pareto Front Upconvert on Multi-objective Building Facility Control Optimization","authors":"Naru Okumura, Tomoaki Takagi, Yoshihiro Ohta, Hiroyuki Sato","doi":"10.1145/3583133.3596339","DOIUrl":"https://doi.org/10.1145/3583133.3596339","url":null,"abstract":"This paper verified the effects of a supervised multi-objective optimization algorithm (SMOA) efficiently upconverting the Pareto front representation by utilizing known solutions on a real-world multi-objective building facility control optimization problem. Also, several sampling methods for evaluating promising candidate solutions in SMOA were proposed and compared. Evolutionary variations, such as crossover and mutation involving randomness, are not preferred in practical scenarios, particularly when the objective functions are computationally expensive. In order to suppress obtaining inferior solutions, SMOA constructs the Pareto front and Pareto set estimation models using known solutions, samples promising candidate solutions, and evaluates them. It was reported that SMOA could efficiently generate well-distributed solutions that upconvert the Pareto front representation compared to evolutionary variations with limited solution evaluations in artificial test problems. This paper focuses on the real-world building facility control problem with 15 known solutions, and results show that SMOA can efficiently improve the Pareto front representation compared to evolutionary variations. Also, results show that crowding distance-based one-time sampling considering the distribution of the known solutions achieved the best Pareto front approximation performance in the sampling methods compared in this paper.","PeriodicalId":422029,"journal":{"name":"Proceedings of the Companion Conference on Genetic and Evolutionary Computation","volume":"400 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123961187","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

On the Behavior of the Mixed-Integer SMS-EMOA on Box-Constrained Quadratic Bi-Objective Models 盒约束二次双目标模型上混合整数SMS-EMOA的行为

Proceedings of the Companion Conference on Genetic and Evolutionary Computation Pub Date : 2023-07-15 DOI: 10.1145/3583133.3596398

O. M. Shir, M. Emmerich

{"title":"On the Behavior of the Mixed-Integer SMS-EMOA on Box-Constrained Quadratic Bi-Objective Models","authors":"O. M. Shir, M. Emmerich","doi":"10.1145/3583133.3596398","DOIUrl":"https://doi.org/10.1145/3583133.3596398","url":null,"abstract":"Existing research lacks studies on how state-of-the-art evolutionary multi-objective algorithms behave when dealing with problems that involve mixed-variable types. To address this gap, we examine how the popular SMS-EMOA performs on a class of problems that involve both continuous and discrete variables. More particularly, we study the algorithmic behavior on a family of mixed-integer (MI) bi-objective optimization problems of partially discretized convex quadratic models. We are considering search by means of a MI Evolution Strategy (ES), and aim to investigate the evolutionary mechanisms as they operate subject to scenarios of box-constrained decision variables. We also account for a white-box approach to solve the models, and qualitatively mention the relative performance of SMS-EMOA with respect to it. We discuss the ES operation within the SMS-EMOA on the current MI case-study, with a particular focus on the step-size adaptation and the success-rate of offspring generation over time. It is evident that progress is made in the initial stage of the optimization, whereas the process tends to stagnate later on. Moreover, for problems whose decision variables are loosely bounded, the step-sizes exhibit effective self-adaptation. We conclude by summarizing the challenges and opportunities when treating MI problems by ES-driven heuristics.","PeriodicalId":422029,"journal":{"name":"Proceedings of the Companion Conference on Genetic and Evolutionary Computation","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124165564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0