Pub Date : 2021-03-24DOI: 10.1504/IJBIC.2021.114101
Jue Li, F. Cao, Honghong Cheng, Yuhua Qian
Convolutional networks bring the performance of many computer vision tasks to unprecedented heights, but at the cost of enormous computation load. To reduce this cost, many model compression tasks have been proposed by eliminating insignificant model structures. For example, convolution filters with small absolute weights are pruned and then fine-tuned to restore reasonable accuracy. However, most of these works rely on pre-trained models without specific analysis of the changes in filters during the training process, resulting in sizable model retraining costs. Different from previous works, we interpret the change of filter behaviour during training from the associated angle, and propose a novel filter pruning method utilising the change rule, which can remove filters with similar functions later in training. According to this strategy, not only can we achieve model compression without fine-tuning, but we can also find a novel perspective to interpret the changing behaviour of the filter during training. Moreover, our approach has been proved to be effective for many advanced CNN architectures.
{"title":"Learning the number of filters in convolutional neural networks","authors":"Jue Li, F. Cao, Honghong Cheng, Yuhua Qian","doi":"10.1504/IJBIC.2021.114101","DOIUrl":"https://doi.org/10.1504/IJBIC.2021.114101","url":null,"abstract":"Convolutional networks bring the performance of many computer vision tasks to unprecedented heights, but at the cost of enormous computation load. To reduce this cost, many model compression tasks have been proposed by eliminating insignificant model structures. For example, convolution filters with small absolute weights are pruned and then fine-tuned to restore reasonable accuracy. However, most of these works rely on pre-trained models without specific analysis of the changes in filters during the training process, resulting in sizable model retraining costs. Different from previous works, we interpret the change of filter behaviour during training from the associated angle, and propose a novel filter pruning method utilising the change rule, which can remove filters with similar functions later in training. According to this strategy, not only can we achieve model compression without fine-tuning, but we can also find a novel perspective to interpret the changing behaviour of the filter during training. Moreover, our approach has been proved to be effective for many advanced CNN architectures.","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"86 1","pages":"75-84"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73054968","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}
Pub Date : 2021-02-23DOI: 10.1504/IJBIC.2021.113356
Jun Suk Kim, C. Ahn
Quantum computing's uniqueness in commencing parallel computation renders unprecedented efficient optimisation as possible. This paper introduces the adaptation of quantum processing to crowding, one of the genetic algorithmic procedures to secure undeveloped individual chromosomes in pursuit of diversifying the target population. We argue that the nature of genetic algorithm to find the best solution in the process of optimisation can be greatly enhanced by the capability of quantum computing to perform multiple computations in parallel. By introducing the relevant quantum mathematics based on Grover's selection algorithm and constructing its mechanism in a quantum simulator, we come to conclusion that our proposed approach is valid in such a way that it can precisely reduce the amount of computation query to finish the crowding process without any impairment in the middle of genetic operations.
{"title":"Expediting population diversification in evolutionary computation with quantum algorithm","authors":"Jun Suk Kim, C. Ahn","doi":"10.1504/IJBIC.2021.113356","DOIUrl":"https://doi.org/10.1504/IJBIC.2021.113356","url":null,"abstract":"Quantum computing's uniqueness in commencing parallel computation renders unprecedented efficient optimisation as possible. This paper introduces the adaptation of quantum processing to crowding, one of the genetic algorithmic procedures to secure undeveloped individual chromosomes in pursuit of diversifying the target population. We argue that the nature of genetic algorithm to find the best solution in the process of optimisation can be greatly enhanced by the capability of quantum computing to perform multiple computations in parallel. By introducing the relevant quantum mathematics based on Grover's selection algorithm and constructing its mechanism in a quantum simulator, we come to conclusion that our proposed approach is valid in such a way that it can precisely reduce the amount of computation query to finish the crowding process without any impairment in the middle of genetic operations.","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"341 1","pages":"63-73"},"PeriodicalIF":0.0,"publicationDate":"2021-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79540023","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}
Pub Date : 2021-02-23DOI: 10.1504/IJBIC.2021.113360
Ü. Çavuşoğlu, A. H. Kökçam
Substitution box (S-box) is one of the most important structures used for byte change operation in block encryption algorithms. An S-box structure with strong cryptological properties makes the encryption algorithm much more resistant to attacks. In this article, a powerful S-box generation algorithm design is presented using genetic algorithm (GA). In the GA-based S-box generation algorithm, the nonlinearity value which is one of the most important S-box evaluation criteria, has been processed. Quality of the generated S-boxes is determined by performance tests. Obtained performance results are compared with the S-boxes in the literature. It has been found that the presented algorithm generates S-boxes with strong cryptological properties.
{"title":"A new approach to design S-box generation algorithm based on genetic algorithm","authors":"Ü. Çavuşoğlu, A. H. Kökçam","doi":"10.1504/IJBIC.2021.113360","DOIUrl":"https://doi.org/10.1504/IJBIC.2021.113360","url":null,"abstract":"Substitution box (S-box) is one of the most important structures used for byte change operation in block encryption algorithms. An S-box structure with strong cryptological properties makes the encryption algorithm much more resistant to attacks. In this article, a powerful S-box generation algorithm design is presented using genetic algorithm (GA). In the GA-based S-box generation algorithm, the nonlinearity value which is one of the most important S-box evaluation criteria, has been processed. Quality of the generated S-boxes is determined by performance tests. Obtained performance results are compared with the S-boxes in the literature. It has been found that the presented algorithm generates S-boxes with strong cryptological properties.","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"1 1","pages":"52-62"},"PeriodicalIF":0.0,"publicationDate":"2021-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87150190","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}
Multi-objective optimisation algorithm based on decomposition (MOEA/D) is a well-known multi-objective optimisation algorithm, which was widely applied for solving multi-objective optimisation problems (MOPs). MOEA/D decomposes a multi-objective problem into a set of scalar single objective sub-problems using aggregation function and evolutionary operator. A further improved version of MOEA/D with dynamic resource allocation strategy (MOEA/D-DRA) has exhibited outstanding performance on CEC2009 in terms of the convergence. However, it is very sensitive to the neighbourhood size. In this paper, a new enchanted MOEA/D-ANA strategy based on the adaptive neighbourhood size adjustment (MOEA/D-ANA) was presented to increase the diversity, which mainly focuses on the solutions density around sub-problems. The experiment results demonstrate that MOEA/D-ANA performs the best compared with other five classical MOEAs on the CEC2009 test instances.
{"title":"Adaptive neighbourhood size adjustment in MOEA/D-DRA","authors":"Meng Xu, Maoqing Zhang, Xingjuan Cai, Guoyou Zhang","doi":"10.1504/IJBIC.2021.113336","DOIUrl":"https://doi.org/10.1504/IJBIC.2021.113336","url":null,"abstract":"Multi-objective optimisation algorithm based on decomposition (MOEA/D) is a well-known multi-objective optimisation algorithm, which was widely applied for solving multi-objective optimisation problems (MOPs). MOEA/D decomposes a multi-objective problem into a set of scalar single objective sub-problems using aggregation function and evolutionary operator. A further improved version of MOEA/D with dynamic resource allocation strategy (MOEA/D-DRA) has exhibited outstanding performance on CEC2009 in terms of the convergence. However, it is very sensitive to the neighbourhood size. In this paper, a new enchanted MOEA/D-ANA strategy based on the adaptive neighbourhood size adjustment (MOEA/D-ANA) was presented to increase the diversity, which mainly focuses on the solutions density around sub-problems. The experiment results demonstrate that MOEA/D-ANA performs the best compared with other five classical MOEAs on the CEC2009 test instances.","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"34 1","pages":"14-23"},"PeriodicalIF":0.0,"publicationDate":"2021-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77242750","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}
Pub Date : 2021-01-01DOI: 10.1504/IJBIC.2021.114088
Jailsingh Bhookya, R. K. Jatoth
{"title":"Sine-cosine-algorithm-based fractional order PID controller tuning for multivariable systems","authors":"Jailsingh Bhookya, R. K. Jatoth","doi":"10.1504/IJBIC.2021.114088","DOIUrl":"https://doi.org/10.1504/IJBIC.2021.114088","url":null,"abstract":"","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"56 1","pages":"113-120"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83935885","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}
Pub Date : 2021-01-01DOI: 10.1504/ijbic.2021.10040610
Hang Wei, Han Huang, Z. Hao, Qinqun Chen, W. Pedrycz, Gang Li
{"title":"A real adjacency matrix-coded evolution algorithm for highly linkage-based routing problems","authors":"Hang Wei, Han Huang, Z. Hao, Qinqun Chen, W. Pedrycz, Gang Li","doi":"10.1504/ijbic.2021.10040610","DOIUrl":"https://doi.org/10.1504/ijbic.2021.10040610","url":null,"abstract":"","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"76 3","pages":"1-12"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91476679","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}
Pub Date : 2021-01-01DOI: 10.1504/ijbic.2021.10040613
S. Konda, L. Panwar, B. K. Panigrahi, Rajesh Kumar, Vishu Gupta
{"title":"Binary fireworks algorithm application for optimal schedule of electric vehicle reserve in traditional and restructured electricity markets","authors":"S. Konda, L. Panwar, B. K. Panigrahi, Rajesh Kumar, Vishu Gupta","doi":"10.1504/ijbic.2021.10040613","DOIUrl":"https://doi.org/10.1504/ijbic.2021.10040613","url":null,"abstract":"","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"69 1","pages":"38-48"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83330701","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}
Pub Date : 2021-01-01DOI: 10.1504/ijbic.2021.10043756
Asma Daoudi, K. Benatchba, Malika Bessedik, Leila Hamdad
{"title":"Performance assessment of biogeography-based multi-objective algorithm for frequency assignment problem","authors":"Asma Daoudi, K. Benatchba, Malika Bessedik, Leila Hamdad","doi":"10.1504/ijbic.2021.10043756","DOIUrl":"https://doi.org/10.1504/ijbic.2021.10043756","url":null,"abstract":"","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"5 1","pages":"199-209"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73487079","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}
Pub Date : 2021-01-01DOI: 10.1504/ijbic.2021.10040609
Yanli Wang, Bo Qu, Jing J. Liang, Yi Hu, Yunpeng Wei
{"title":"Research on the ensemble feature selection algorithm based on multimodal optimisation techniques","authors":"Yanli Wang, Bo Qu, Jing J. Liang, Yi Hu, Yunpeng Wei","doi":"10.1504/ijbic.2021.10040609","DOIUrl":"https://doi.org/10.1504/ijbic.2021.10040609","url":null,"abstract":"","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"31 1","pages":"49-58"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82461047","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}
Pub Date : 2021-01-01DOI: 10.1504/IJBIC.2021.114873
Zhaolu Guo, Wensheng Zhang, Shenwen Wang
{"title":"Improved gravitational search algorithm based on chaotic local search","authors":"Zhaolu Guo, Wensheng Zhang, Shenwen Wang","doi":"10.1504/IJBIC.2021.114873","DOIUrl":"https://doi.org/10.1504/IJBIC.2021.114873","url":null,"abstract":"","PeriodicalId":13636,"journal":{"name":"Int. J. Bio Inspired Comput.","volume":"52 1","pages":"154-164"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90797196","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}
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：