Pub Date : 2023-11-22DOI: 10.1007/s10710-023-09467-x
W. B. Langdon
It is 30 years since John R. Koza published “Jaws”, the first book on genetic programming [Genetic Programming: On the Programming of Computers by Means of Natural Selection. MIT Press (1992)]. I recount and expand the celebration at GECCO 2022, very briefly summarise some of what the rest of us have done and make suggestions for the next thirty years of GP research.
30年前，John R. Koza出版了第一本关于遗传编程的书《大白鲨》(genetic programming: on the programming of Computers by Means of Natural Selection)。麻省理工学院出版社(1992)。我叙述并扩展了GECCO 2022的庆祝活动，非常简要地总结了我们其他人所做的一些工作，并为未来三十年的GP研究提出了建议。
{"title":"Jaws 30","authors":"W. B. Langdon","doi":"10.1007/s10710-023-09467-x","DOIUrl":"https://doi.org/10.1007/s10710-023-09467-x","url":null,"abstract":"<p>It is 30 years since John R. Koza published “Jaws”, the first book on genetic programming [Genetic Programming: On the Programming of Computers by Means of Natural Selection. MIT Press (1992)]. I recount and expand the celebration at GECCO 2022, very briefly summarise some of what the rest of us have done and make suggestions for the next thirty years of GP research.</p>","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"25 2","pages":""},"PeriodicalIF":2.6,"publicationDate":"2023-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138496062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-22DOI: 10.1007/s10710-023-09474-y
W. B. Langdon
{"title":"Response to comments on “Jaws 30”","authors":"W. B. Langdon","doi":"10.1007/s10710-023-09474-y","DOIUrl":"https://doi.org/10.1007/s10710-023-09474-y","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"25 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2023-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138496060","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-22DOI: 10.1007/s10710-023-09469-9
Jason H. Moore
{"title":"Is the evolution metaphor still necessary or even useful for genetic programming?","authors":"Jason H. Moore","doi":"10.1007/s10710-023-09469-9","DOIUrl":"https://doi.org/10.1007/s10710-023-09469-9","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"71 3","pages":"1-3"},"PeriodicalIF":2.6,"publicationDate":"2023-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139246869","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-22DOI: 10.1007/s10710-023-09470-2
Colin G. Johnson
{"title":"New directions in fitness evaluation: commentary on Langdon’s JAWS30","authors":"Colin G. Johnson","doi":"10.1007/s10710-023-09470-2","DOIUrl":"https://doi.org/10.1007/s10710-023-09470-2","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"2 3","pages":"1-3"},"PeriodicalIF":2.6,"publicationDate":"2023-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139247224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-22DOI: 10.1007/s10710-023-09473-z
Malcolm I. Heywood
At the 30th anniversary of ‘Jaws’, the Genetic programming field has much to celebrate. However, in order continue to build on these successes, it might be necessary to look more deeply into the “less successful” and/or “less explored” topics. We consider the role of FPGA and GPU platforms from the former and coevolution from the latter.
{"title":"W. B. Langdon “Jaws 30”","authors":"Malcolm I. Heywood","doi":"10.1007/s10710-023-09473-z","DOIUrl":"https://doi.org/10.1007/s10710-023-09473-z","url":null,"abstract":"<p>At the 30th anniversary of ‘Jaws’, the Genetic programming field has much to celebrate. However, in order continue to build on these successes, it might be necessary to look more deeply into the “less successful” and/or “less explored” topics. We consider the role of <i>FPGA and GPU platforms</i> from the former and <i>coevolution</i> from the latter.</p>","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"25 S2","pages":""},"PeriodicalIF":2.6,"publicationDate":"2023-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138496061","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-08DOI: 10.1007/s10710-023-09462-2
David Wittenberg, Franz Rothlauf, Christian Gagné
Abstract Denoising autoencoder genetic programming (DAE-GP) is a novel neural network-based estimation of distribution genetic programming approach that uses denoising autoencoder long short-term memory networks as a probabilistic model to replace the standard mutation and recombination operators of genetic programming. At each generation, the idea is to capture promising properties of the parent population in a probabilistic model and to use corruption to transfer variations of these properties to the offspring. This work studies the influence of corruption and sampling steps on search. Corruption partially mutates candidate solutions that are used as input to the model, whereas the number of sampling steps defines how often we re-use the output during model sampling as input to the model. We study the generalization of the royal tree problem, the Airfoil problem, and the Pagie-1 problem, and find that both corruption strength and the number of sampling steps influence exploration and exploitation in search and affect performance: exploration increases with stronger corruption and lower number of sampling steps. The results indicate that both corruption and sampling steps are key to the success of the DAE-GP: it permits us to balance the exploration and exploitation behavior in search, resulting in an improved search quality. However, also selection is important for exploration and exploitation and should be chosen wisely.
{"title":"Denoising autoencoder genetic programming: strategies to control exploration and exploitation in search","authors":"David Wittenberg, Franz Rothlauf, Christian Gagné","doi":"10.1007/s10710-023-09462-2","DOIUrl":"https://doi.org/10.1007/s10710-023-09462-2","url":null,"abstract":"Abstract Denoising autoencoder genetic programming (DAE-GP) is a novel neural network-based estimation of distribution genetic programming approach that uses denoising autoencoder long short-term memory networks as a probabilistic model to replace the standard mutation and recombination operators of genetic programming. At each generation, the idea is to capture promising properties of the parent population in a probabilistic model and to use corruption to transfer variations of these properties to the offspring. This work studies the influence of corruption and sampling steps on search. Corruption partially mutates candidate solutions that are used as input to the model, whereas the number of sampling steps defines how often we re-use the output during model sampling as input to the model. We study the generalization of the royal tree problem, the Airfoil problem, and the Pagie-1 problem, and find that both corruption strength and the number of sampling steps influence exploration and exploitation in search and affect performance: exploration increases with stronger corruption and lower number of sampling steps. The results indicate that both corruption and sampling steps are key to the success of the DAE-GP: it permits us to balance the exploration and exploitation behavior in search, resulting in an improved search quality. However, also selection is important for exploration and exploitation and should be chosen wisely.","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":" 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135340710","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Geometric semantic genetic programming (GSGP) is a popular form of GP where the effect of crossover and mutation can be expressed as geometric operations on a semantic space. A recent study showed that GSGP can be hybridized with a standard gradient-based optimized, Adam, commonly used in training artificial neural networks.We expand upon that work by considering more gradient-based optimizers, a deeper investigation of their parameters, how the hybridization is performed, and a more comprehensive set of benchmark problems. With the correct choice of hyperparameters, this hybridization improves the performances of GSGP and allows it to reach the same fitness values with fewer fitness evaluations.
{"title":"On the hybridization of geometric semantic GP with gradient-based optimizers","authors":"Gloria Pietropolli, Luca Manzoni, Alessia Paoletti, Mauro Castelli","doi":"10.1007/s10710-023-09463-1","DOIUrl":"https://doi.org/10.1007/s10710-023-09463-1","url":null,"abstract":"Abstract Geometric semantic genetic programming (GSGP) is a popular form of GP where the effect of crossover and mutation can be expressed as geometric operations on a semantic space. A recent study showed that GSGP can be hybridized with a standard gradient-based optimized, Adam, commonly used in training artificial neural networks.We expand upon that work by considering more gradient-based optimizers, a deeper investigation of their parameters, how the hybridization is performed, and a more comprehensive set of benchmark problems. With the correct choice of hyperparameters, this hybridization improves the performances of GSGP and allows it to reach the same fitness values with fewer fitness evaluations.","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"20 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136158712","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-26DOI: 10.1007/s10710-023-09464-0
Illya Bakurov, Marco Buzzelli, Raimondo Schettini, Mauro Castelli, Leonardo Vanneschi
Abstract Semantic segmentation consists of classifying each pixel of an image and constitutes an essential step towards scene recognition and understanding. Deep convolutional encoder–decoder neural networks now constitute state-of-the-art methods in the field of semantic segmentation. The problem of street scenes’ segmentation for automotive applications constitutes an important application field of such networks and introduces a set of imperative exigencies. Since the models need to be executed on self-driving vehicles to make fast decisions in response to a constantly changing environment, they are not only expected to operate reliably but also to process the input images rapidly. In this paper, we explore genetic programming (GP) as a meta-model that combines four different efficiency-oriented networks for the analysis of urban scenes. Notably, we present and examine two approaches. In the first approach, we represent solutions as GP trees that combine networks’ outputs such that each output class’s prediction is obtained through the same meta-model. In the second approach, we propose representing solutions as lists of GP trees, each designed to provide a unique meta-model for a given target class. The main objective is to develop efficient and accurate combination models that could be easily interpreted, therefore allowing gathering some hints on how to improve the existing networks. The experiments performed on the Cityscapes dataset of urban scene images with semantic pixel-wise annotations confirm the effectiveness of the proposed approach. Specifically, our best-performing models improve systems’ generalization ability by approximately 5% compared to traditional ensembles, 30% for the less performing state-of-the-art CNN and show competitive results with respect to state-of-the-art ensembles. Additionally, they are small in size, allow interpretability, and use fewer features due to GP’s automatic feature selection.
{"title":"Semantic segmentation network stacking with genetic programming","authors":"Illya Bakurov, Marco Buzzelli, Raimondo Schettini, Mauro Castelli, Leonardo Vanneschi","doi":"10.1007/s10710-023-09464-0","DOIUrl":"https://doi.org/10.1007/s10710-023-09464-0","url":null,"abstract":"Abstract Semantic segmentation consists of classifying each pixel of an image and constitutes an essential step towards scene recognition and understanding. Deep convolutional encoder–decoder neural networks now constitute state-of-the-art methods in the field of semantic segmentation. The problem of street scenes’ segmentation for automotive applications constitutes an important application field of such networks and introduces a set of imperative exigencies. Since the models need to be executed on self-driving vehicles to make fast decisions in response to a constantly changing environment, they are not only expected to operate reliably but also to process the input images rapidly. In this paper, we explore genetic programming (GP) as a meta-model that combines four different efficiency-oriented networks for the analysis of urban scenes. Notably, we present and examine two approaches. In the first approach, we represent solutions as GP trees that combine networks’ outputs such that each output class’s prediction is obtained through the same meta-model. In the second approach, we propose representing solutions as lists of GP trees, each designed to provide a unique meta-model for a given target class. The main objective is to develop efficient and accurate combination models that could be easily interpreted, therefore allowing gathering some hints on how to improve the existing networks. The experiments performed on the Cityscapes dataset of urban scene images with semantic pixel-wise annotations confirm the effectiveness of the proposed approach. Specifically, our best-performing models improve systems’ generalization ability by approximately 5% compared to traditional ensembles, 30% for the less performing state-of-the-art CNN and show competitive results with respect to state-of-the-art ensembles. Additionally, they are small in size, allow interpretability, and use fewer features due to GP’s automatic feature selection.","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"32 9","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134908280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-20DOI: 10.1007/s10710-023-09461-3
Fabrício Olivetti de França
{"title":"Alleviating overfitting in transformation-interaction-rational symbolic regression with multi-objective optimization","authors":"Fabrício Olivetti de França","doi":"10.1007/s10710-023-09461-3","DOIUrl":"https://doi.org/10.1007/s10710-023-09461-3","url":null,"abstract":"","PeriodicalId":50424,"journal":{"name":"Genetic Programming and Evolvable Machines","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135569369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：