Ansel Y. Rodríguez González, Ramón Aranda, Miguel A. Alvarez-Carmona, Y. Martínez-López, J. M. Quintana
{"title":"Applying ring cellular encode-decode UMDA to risk-based energy scheduling","authors":"Ansel Y. Rodríguez González, Ramón Aranda, Miguel A. Alvarez-Carmona, Y. Martínez-López, J. M. Quintana","doi":"10.1145/3520304.3534055","DOIUrl":"https://doi.org/10.1145/3520304.3534055","url":null,"abstract":"One of the challenges to making cities and human settlements inclusive, safe, resilient, and sustainable is to manage energy efficiently. Thus, developing algorithms to address smart grid management-related problems is still an open research area. In this context, the \"Competition on Evolutionary Computation in the Energy Domain\", launched in 2017, proposes the risk-based energy scheduling problem in the 2022 edition. In this paper, the algorithm RCED-UMDA is applied to solve the risk-based energy scheduling. Experimental results reveal that RCED-UMDA achieves better behavior than the top three algorithms of each track of the past 2021 competition edition.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115595669","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}
{"title":"A distributed particle swarm optimization algorithm for distributed clustering","authors":"Ziyan Li, X. Guo, Wei-neng Chen, Xiao-Min Hu","doi":"10.1145/3520304.3529016","DOIUrl":"https://doi.org/10.1145/3520304.3529016","url":null,"abstract":"With the rapid development of the Internet of Things, distributed clustering has emerged as an attractive research topic. Although evolutionary computation (EC) has been introduced as an effective approach to solve complex data clustering problems, few studies realize the use of EC for distributed clustering. In this paper, we intend to propose a distributed particle swarm optimization (PSO) for distributed clustering. The main idea of the work is to take advantage of the natural distributed computing feature of EC to solve distributed clustering. Since each site only knows a part of the data, the performance of distributed clustering algorithm usually deteriorates. To address this challenge, the proposed algorithm includes two main steps. Firstly, PSO is introduced in each local site to perform local data clustering. Secondly, to combine the local clustering results together, the K-means algorithm is used in the global site to generate the global clustering result. Experimental results demonstrate the overall performance superiority of the proposed algorithm compared with two other algorithms.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"1086 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116037474","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}
{"title":"The case for grammatical evolution in test generation","authors":"Aidan Murphy, Thomas Laurent, Anthony Ventresque","doi":"10.1145/3520304.3534042","DOIUrl":"https://doi.org/10.1145/3520304.3534042","url":null,"abstract":"Generating tests for software is an important, but difficult, task. Search-based test generation is promising, as it reduces the time required from human experts, but suffers from many problems and limitations. Namely, the inability to fully incorporate a tester's domain knowledge into the search, its difficulty in creating very complex objects, and the problems associated with variable length tests. This paper illustrates how Grammatical Evolution could address and provide a possible solution to each of these concerns.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116324316","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}
{"title":"Evolutionary training of deep neural networks on heterogeneous computing environments","authors":"Subodh Kalia, C. Mohan, R. Nemani","doi":"10.1145/3520304.3533954","DOIUrl":"https://doi.org/10.1145/3520304.3533954","url":null,"abstract":"Deep neural networks are typically trained using gradient-based optimizers such as error backpropagation. This study proposes a framework based on Evolutionary Algorithms (EAs) to train deep neural networks without gradients. The network parameters, which may vary up to millions, are considered optimization variables. We demonstrate the training of an encoder-decoder segmentation network (U-Net) and Long Short-Term Memory (LSTM) model using (μ + λ)-ES, Genetic Algorithm, and Particle Swarm Optimization. The framework can train models with forward propagation on machines with different hardware in a cluster computing environment. We compare prediction results from the two models trained using our framework and backpropagation. We show that the neural networks can be trained in less time on CPUs as compared to the training on specialized compute-intensive GPUs.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123955616","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}
{"title":"A surrogate model-based genetic algorithm for the optimal policy in cart-pole balancing environments","authors":"Seung-Soo Shin, Yong-Hyuk Kim","doi":"10.1145/3520304.3528788","DOIUrl":"https://doi.org/10.1145/3520304.3528788","url":null,"abstract":"Agent's learning for completing a given mission needs a lot of trial-and-error. Generally, an agent's performance is evaluated by a simulation. Thus, exorbitant simulation costs may be incurred when an agent is trained to solve real-world problems. We propose a surrogate model-based genetic algorithm for reducing simulation-based evaluation costs of the cart-pole balancing problem. Policy networks and returns by the networks were collected by training a deep Q-network (DQN) among reinforcement learning methods. A surrogate model for the fitness function of a genetic algorithm was completed through our heuristic measure and DNN-based models trained on DQN data. Our surrogate model-based genetic algorithm performed much faster than that with simulation-based evaluation, and it could successfully find optimal solutions that achieves the maximum cumulative reward.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125051321","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}
{"title":"Statistical analyses for multi-objective stochastic optimization algorithms: GECCO 2022 tutorial","authors":"T. Eftimov, P. Korošec","doi":"10.1145/3520304.3533668","DOIUrl":"https://doi.org/10.1145/3520304.3533668","url":null,"abstract":"","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128334847","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}
{"title":"Automatically exploring computer system design spaces","authors":"Bobby R. Bruce","doi":"10.1145/3520304.3534021","DOIUrl":"https://doi.org/10.1145/3520304.3534021","url":null,"abstract":"While much research has focused on using search to optimize software, it is possible to use these same approaches to optimize other parts of the computer system. With decreased costs in silicon customization, and the return of centralized systems carrying out specialized tasks, the time is right to begin thinking about tools to optimize systems for the needs of specific software targets. In the approach outlined, the standard Genetic Improvement process is flipped with source code considered static and the remainder of the computer system altered to the needs of the software. The project proposed is preliminary research into incorporating grammar-based GP with an advanced computer architecture simulator to automatically design computer systems. I argue this approach has the potential to significantly improve the design of computer systems while reducing manual effort.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129109628","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}
Sanjai Pathak, Ashish Mani, Mayank Sharma, A. Chatterjee
{"title":"Rethinking of controller placement problem from static optimization to multi-objective dynamic optimization","authors":"Sanjai Pathak, Ashish Mani, Mayank Sharma, A. Chatterjee","doi":"10.1145/3520304.3528929","DOIUrl":"https://doi.org/10.1145/3520304.3528929","url":null,"abstract":"The controller placement problem (CPP) is modeled as a dynamic multi-objective combinatorial optimization problem (DMOCPP). A novel algorithm is introduced - Dynamic multi-objective controller placement algorithm (DMOCPA) based on the multi-population and multi-objective quantum-inspired Salp Swarm Algorithm (MMQSSA) to solve the DMOCPP. The proposed work advances the field of traditional networking, to address the issues when the networks are highly dynamic and complex.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"168 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125662612","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}
{"title":"Incorporating sub-programs as knowledge in program synthesis by PushGP and adaptive replacement mutation","authors":"Yifan He, C. Aranha, T. Sakurai","doi":"10.1145/3520304.3528891","DOIUrl":"https://doi.org/10.1145/3520304.3528891","url":null,"abstract":"Program synthesis aims to build an intelligent agent that composes computer programs to solve problems. Genetic programming (GP) provides an evolutionary solution for the program synthesis task. A typical GP includes a random initialization, an unguided variation, and a fitness-guided selection to search for a solution program. However, several recent studies have shown the importance of using prior knowledge in different components of the GP. This study investigates the effectiveness of incorporating sub-programs as \"prior knowledge\" into the variation process of GP by Replacement Mutation. We further design an adaptive strategy that allows the automatic selection of the helpful sub-programs to the search process from an archive (including helpful and unhelpful ones). With handcrafted sub-program archives, we verify the effectiveness of the Adaptive Replacement Mutation method in success rate. We demonstrate the effectiveness of our approach with transferred archives on two composite problems.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130224153","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}
{"title":"Generative hyper-heuristics","authors":"D. Tauritz, J. Woodward","doi":"10.1145/3520304.3533646","DOIUrl":"https://doi.org/10.1145/3520304.3533646","url":null,"abstract":"","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131663625","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}