A meme in the context of optimization represents a unit of algorithmic abstraction that dictates how solution search is carried out. At a higher level, a meta-meme serves as an encapsulation of the scheme of interplay between memes involved in the search process. This paper puts forth the notion of neural meta-memes to extend the collective capacity of memes in problem-solving. We term this as Neural Meta-Memes Framework (NMMF) for combinatorial optimization. NMMF models basic optimization algorithms as memes and manages them dynamically. We show the efficacy of the proposed NMMF through empirical study on a class of combinatorial optimization problem, the quadratic assignment problem (QAP).
{"title":"Neural meta-memes framework for managing search algorithms in combinatorial optimization","authors":"L. Song, M. Lim, Y. Ong","doi":"10.1109/MC.2011.5953634","DOIUrl":"https://doi.org/10.1109/MC.2011.5953634","url":null,"abstract":"A meme in the context of optimization represents a unit of algorithmic abstraction that dictates how solution search is carried out. At a higher level, a meta-meme serves as an encapsulation of the scheme of interplay between memes involved in the search process. This paper puts forth the notion of neural meta-memes to extend the collective capacity of memes in problem-solving. We term this as Neural Meta-Memes Framework (NMMF) for combinatorial optimization. NMMF models basic optimization algorithms as memes and manages them dynamically. We show the efficacy of the proposed NMMF through empirical study on a class of combinatorial optimization problem, the quadratic assignment problem (QAP).","PeriodicalId":441186,"journal":{"name":"2011 IEEE Workshop on Memetic Computing (MC)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114194261","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}
Cluster expansion provides a powerful tool in materials modeling. It has enabled an efficient prediction of the atomic properties of materials with the combination of the modern quantum calculation theory. To construct an accurate cluster expansion model, a few important cluster figures should be identified. This paper proposes a novel figure selection method based on memetic algorithm (MA), which is a synergy of genetic algorithm (GA) and orthogonal matching pursuit (OMP) based memetic operation. The memetic operation is designed to fine-tunes the solutions of GA and accelerate the convergence of the search. The performance of the proposed method is evaluated on two binary alloy datasets. Comparative study to other state-of-the-art figure selection methods demonstrates that the proposed method is capable of obtaining better or competitive prediction accuracy and searching the figure space efficiently.
{"title":"Memetic figure selection for cluster expansion in binary alloy systems","authors":"Zexuan Zhu, Z. Ji, Xiaofeng Fan, J. Kuo","doi":"10.1109/MC.2011.5953635","DOIUrl":"https://doi.org/10.1109/MC.2011.5953635","url":null,"abstract":"Cluster expansion provides a powerful tool in materials modeling. It has enabled an efficient prediction of the atomic properties of materials with the combination of the modern quantum calculation theory. To construct an accurate cluster expansion model, a few important cluster figures should be identified. This paper proposes a novel figure selection method based on memetic algorithm (MA), which is a synergy of genetic algorithm (GA) and orthogonal matching pursuit (OMP) based memetic operation. The memetic operation is designed to fine-tunes the solutions of GA and accelerate the convergence of the search. The performance of the proposed method is evaluated on two binary alloy datasets. Comparative study to other state-of-the-art figure selection methods demonstrates that the proposed method is capable of obtaining better or competitive prediction accuracy and searching the figure space efficiently.","PeriodicalId":441186,"journal":{"name":"2011 IEEE Workshop on Memetic Computing (MC)","volume":"115 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114249138","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}
Jiarui Zhou, Z. Ji, L. Shen, Zexuan Zhu, Siping Chen
A Gabor filters based face recognition algorithm named POMA-Gabor is proposed in this paper. The algorithm uses particular Gabor wavelets in the feature extraction on specific areas of the face image and a particle swarm optimization (PSO) based memetic algorithm (POMA), which combines comprehensive learning particle swarm optimizer (CLPSO) global search and self-adaptive intelligent single particle optimizer (AdpISPO) local search, is introduced to select the Gabor filter parameters. The experimental results demonstrate that POMA obtains better performance than other comparative PSO algorithms. Employing POMA for Gabor filter design, POMA-Gabor is capable of obtaining more representative information and higher recognition rate with less computational time.
{"title":"PSO based memetic algorithm for face recognition Gabor filters selection","authors":"Jiarui Zhou, Z. Ji, L. Shen, Zexuan Zhu, Siping Chen","doi":"10.1109/MC.2011.5953631","DOIUrl":"https://doi.org/10.1109/MC.2011.5953631","url":null,"abstract":"A Gabor filters based face recognition algorithm named POMA-Gabor is proposed in this paper. The algorithm uses particular Gabor wavelets in the feature extraction on specific areas of the face image and a particle swarm optimization (PSO) based memetic algorithm (POMA), which combines comprehensive learning particle swarm optimizer (CLPSO) global search and self-adaptive intelligent single particle optimizer (AdpISPO) local search, is introduced to select the Gabor filter parameters. The experimental results demonstrate that POMA obtains better performance than other comparative PSO algorithms. Employing POMA for Gabor filter design, POMA-Gabor is capable of obtaining more representative information and higher recognition rate with less computational time.","PeriodicalId":441186,"journal":{"name":"2011 IEEE Workshop on Memetic Computing (MC)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115139550","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}
A novel Hybrid Algorithm called Hybrid Algorithm based on Differential Immune Clone with Orthogonal design method (OHADIC) is proposed in this paper, which can avoid the decrease of population diversity and accelerate the convergence rate in evolutionary process. The novel algorithm adopts several main operators to evolve two populations; they are clone reproduction and selection, differential mutation, crossover and selection. Moreover, the orthogonal design method is not only to be used to design orthogonal crossover, but also is adapted to scheme orthogonal local search. In experiments, a wide range of benchmark functions is used to validate the novel hybrid algorithm. Performance comparisons with other well-known differential evolution algorithms including DE, JADE and SADE are also presented, and it is shown that OHADIC has better performance in optimizing these functions.
{"title":"Hybrid Algorithm based on Differential Immune Clone with Orthogonal design method","authors":"Wenping Ma, Feifei Ti, Maoguo Gong","doi":"10.1109/MC.2011.5953630","DOIUrl":"https://doi.org/10.1109/MC.2011.5953630","url":null,"abstract":"A novel Hybrid Algorithm called Hybrid Algorithm based on Differential Immune Clone with Orthogonal design method (OHADIC) is proposed in this paper, which can avoid the decrease of population diversity and accelerate the convergence rate in evolutionary process. The novel algorithm adopts several main operators to evolve two populations; they are clone reproduction and selection, differential mutation, crossover and selection. Moreover, the orthogonal design method is not only to be used to design orthogonal crossover, but also is adapted to scheme orthogonal local search. In experiments, a wide range of benchmark functions is used to validate the novel hybrid algorithm. Performance comparisons with other well-known differential evolution algorithms including DE, JADE and SADE are also presented, and it is shown that OHADIC has better performance in optimizing these functions.","PeriodicalId":441186,"journal":{"name":"2011 IEEE Workshop on Memetic Computing (MC)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123998557","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}
Yan Li, Lai Jiang, Yan Yin, Fangfang Liu, Hang Yu, Zhen Ji
ETC (Electronic Tolling Collection) systems develop rapidly in China in order to relieve the traffic congestion. As the key component of the ETC system, the design of the Radio Frequency (RF) transceiver is usually a tedious and experienced based work due to its high non-linearity. An automatic system level optimization method based on Memetic algorithm (MA) is proposed in this paper. A fitness function describing the relationship between receiver output signal-to-noise ratio (SNRout) and 9 system level parameters was derived and was optimized by the MA method. The correctness of the MA method was verified by the ADS simulation. A close result was obtained and the effects of the 9 parameters on the SNRout were discussed. The future design of the whole transceiver system can be based on this method.
{"title":"System optimization of a 5.8 GHz ETC receiver using Memetic algorithm","authors":"Yan Li, Lai Jiang, Yan Yin, Fangfang Liu, Hang Yu, Zhen Ji","doi":"10.1109/MC.2011.5953636","DOIUrl":"https://doi.org/10.1109/MC.2011.5953636","url":null,"abstract":"ETC (Electronic Tolling Collection) systems develop rapidly in China in order to relieve the traffic congestion. As the key component of the ETC system, the design of the Radio Frequency (RF) transceiver is usually a tedious and experienced based work due to its high non-linearity. An automatic system level optimization method based on Memetic algorithm (MA) is proposed in this paper. A fitness function describing the relationship between receiver output signal-to-noise ratio (SNRout) and 9 system level parameters was derived and was optimized by the MA method. The correctness of the MA method was verified by the ADS simulation. A close result was obtained and the effects of the 9 parameters on the SNRout were discussed. The future design of the whole transceiver system can be based on this method.","PeriodicalId":441186,"journal":{"name":"2011 IEEE Workshop on Memetic Computing (MC)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134338691","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}
Giovanni Iacca, R. Mallipeddi, E. Mininno, Ferrante Neri, P. N. Suganthan
Although Differential Evolution is an efficient and versatile optimizer, it has a wide margin of improvement. During the latest years much effort of computer scientists studying Differential Evolution has been oriented towards the improvement of the algorithmic paradigm by adding and modifying components. In particular, two modifications lead to important improvements to the original algorithmic performance. The first is the super-fit mechanism, that is the injection at the beginning of the optimization process of a solution previously improved by another algorithm. The second is the progressive reduction of the population size during the evolution of the population. Recently, the algorithmic paradigm of compact Differential Evolution has been introduced. This class of algorithm does not process a population of solutions but its probabilistic representation. In this way, the Differential Evolution can be employed on a device characterized by a limited memory, such as microcontroller or a Graphics Processing Unit. This paper proposes the implementation of the two modifications mentioned above in the context of compact optimization. The compact versions of memetic super-fit mechanism and population size reduction have been tested in this paper and their benefits highlighted. The main finding of this paper is that although separately these modifications do not robustly lead to significant performance improvements, the combined action of the two mechanism appears to be extremely efficient in compact optimization. The resulting algorithm succeeds at handling very diverse fitness landscapes and appears to improve on a regular basis the performance of a standard compact Differential Evolution.
{"title":"Super-fit and population size reduction in compact Differential Evolution","authors":"Giovanni Iacca, R. Mallipeddi, E. Mininno, Ferrante Neri, P. N. Suganthan","doi":"10.1109/MC.2011.5953633","DOIUrl":"https://doi.org/10.1109/MC.2011.5953633","url":null,"abstract":"Although Differential Evolution is an efficient and versatile optimizer, it has a wide margin of improvement. During the latest years much effort of computer scientists studying Differential Evolution has been oriented towards the improvement of the algorithmic paradigm by adding and modifying components. In particular, two modifications lead to important improvements to the original algorithmic performance. The first is the super-fit mechanism, that is the injection at the beginning of the optimization process of a solution previously improved by another algorithm. The second is the progressive reduction of the population size during the evolution of the population. Recently, the algorithmic paradigm of compact Differential Evolution has been introduced. This class of algorithm does not process a population of solutions but its probabilistic representation. In this way, the Differential Evolution can be employed on a device characterized by a limited memory, such as microcontroller or a Graphics Processing Unit. This paper proposes the implementation of the two modifications mentioned above in the context of compact optimization. The compact versions of memetic super-fit mechanism and population size reduction have been tested in this paper and their benefits highlighted. The main finding of this paper is that although separately these modifications do not robustly lead to significant performance improvements, the combined action of the two mechanism appears to be extremely efficient in compact optimization. The resulting algorithm succeeds at handling very diverse fitness landscapes and appears to improve on a regular basis the performance of a standard compact Differential Evolution.","PeriodicalId":441186,"journal":{"name":"2011 IEEE Workshop on Memetic Computing (MC)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126343456","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}
In this paper, a novel immune algorithm with dynamic memetic Cauchy mutation (DMCMIA) for multi-objective optimization is proposed. The idea of memetics is incorporated into the mutation process and a dynamic memetic Cauchy mutation (DMCM) operator is developed. The DMCM operator combines global exploration and local refinement efficiently, which adopts a generation-dependent parameter to guarantee a good balance between global search and local search. Comparison is made to another multi-objective optimization algorithm, nondominated neighbor immune algorithm, termed as NNIA, in solving five ZDT and five DTLZ standard test problems. Simulation results based on coverage of two set, convergence metric and spacing show that DMCMIA performs better than NNIA in generating approximations to the true Pareto front. In addition, the effectiveness of the novel dynamic memetic Cauchy mutation is verified by comparison to polynomial mutation and Gaussian mutation, the experimental results reinforce the advantage of the DMCM operator.
{"title":"Multi-objective immune algorithm with dynamic memetic Cauchy mutation","authors":"Yanli Yang, Hanbing Fang","doi":"10.1109/MC.2011.5953629","DOIUrl":"https://doi.org/10.1109/MC.2011.5953629","url":null,"abstract":"In this paper, a novel immune algorithm with dynamic memetic Cauchy mutation (DMCMIA) for multi-objective optimization is proposed. The idea of memetics is incorporated into the mutation process and a dynamic memetic Cauchy mutation (DMCM) operator is developed. The DMCM operator combines global exploration and local refinement efficiently, which adopts a generation-dependent parameter to guarantee a good balance between global search and local search. Comparison is made to another multi-objective optimization algorithm, nondominated neighbor immune algorithm, termed as NNIA, in solving five ZDT and five DTLZ standard test problems. Simulation results based on coverage of two set, convergence metric and spacing show that DMCMIA performs better than NNIA in generating approximations to the true Pareto front. In addition, the effectiveness of the novel dynamic memetic Cauchy mutation is verified by comparison to polynomial mutation and Gaussian mutation, the experimental results reinforce the advantage of the DMCM operator.","PeriodicalId":441186,"journal":{"name":"2011 IEEE Workshop on Memetic Computing (MC)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126351252","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：