Pub Date : 2013-06-20DOI: 10.1109/CEC.2013.6557557
D. Molina, Amilkar Puris, Rafael Bello, F. Herrera
Many real-world problems have several optima, and the aim of niching optimisation algorithms is to obtain the different global optima, and not only the best solution. One common technique to create niches is the clearing method that removes solutions too close to better ones. Unfortunately, clearing is very sensitive to the niche radius, and its right value depends on the problem (in real-world problems the minimum distance between optima is unknown). In this work we propose a niching algorithm that uses clearing with an adaptive niche radius, that decreases during the run. The proposal uses an external memory that stores current global optima to avoid losing found optima during the clearing process, allowing a non-elitist search. This algorithm applies this clearing method to a mesh of solutions, expanded by the generation of nodes using combination methods between the nodes, their best neighbour, and their nearest current global optima in the population (current global optima are nodes with fitness very similar to current best fitness). The proposal is tested on the competition benchmark proposed in the Special Session Niching Methods for Multimodal Optimization, and compared with other algorithms. The proposal obtains very good results detecting global optima. In comparisons with other algorithm, this proposal obtains the best results, proving to be a very competitive niching algorithm.
{"title":"Variable mesh optimization for the 2013 CEC Special Session Niching Methods for Multimodal Optimization","authors":"D. Molina, Amilkar Puris, Rafael Bello, F. Herrera","doi":"10.1109/CEC.2013.6557557","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557557","url":null,"abstract":"Many real-world problems have several optima, and the aim of niching optimisation algorithms is to obtain the different global optima, and not only the best solution. One common technique to create niches is the clearing method that removes solutions too close to better ones. Unfortunately, clearing is very sensitive to the niche radius, and its right value depends on the problem (in real-world problems the minimum distance between optima is unknown). In this work we propose a niching algorithm that uses clearing with an adaptive niche radius, that decreases during the run. The proposal uses an external memory that stores current global optima to avoid losing found optima during the clearing process, allowing a non-elitist search. This algorithm applies this clearing method to a mesh of solutions, expanded by the generation of nodes using combination methods between the nodes, their best neighbour, and their nearest current global optima in the population (current global optima are nodes with fitness very similar to current best fitness). The proposal is tested on the competition benchmark proposed in the Special Session Niching Methods for Multimodal Optimization, and compared with other algorithms. The proposal obtains very good results detecting global optima. In comparisons with other algorithm, this proposal obtains the best results, proving to be a very competitive niching algorithm.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123895546","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 : 2013-06-20DOI: 10.1109/CEC.2013.6557624
A. Blair
We describe a new programming language called HERCL, designed for evolutionary computation with the specific aim of allowing new programs to be created by combining patches of code from different parts of other programs, at multiple scales. Large-scale patches are followed up by smaller-scale patches or mutations, recursively, to produce a global random search strategy known as hierarchical evolutionary re-combination. We demonstrate the proposed system on the task of learning to encode with the Caesar or Vigenere Cipher, and show how the evolution of one task may fruitfully be cross-pollinated with evolved solutions from other related tasks.
{"title":"Learning the Caesar and Vigenere Cipher by hierarchical evolutionary re-combination","authors":"A. Blair","doi":"10.1109/CEC.2013.6557624","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557624","url":null,"abstract":"We describe a new programming language called HERCL, designed for evolutionary computation with the specific aim of allowing new programs to be created by combining patches of code from different parts of other programs, at multiple scales. Large-scale patches are followed up by smaller-scale patches or mutations, recursively, to produce a global random search strategy known as hierarchical evolutionary re-combination. We demonstrate the proposed system on the task of learning to encode with the Caesar or Vigenere Cipher, and show how the evolution of one task may fruitfully be cross-pollinated with evolved solutions from other related tasks.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123651313","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 : 2013-06-20DOI: 10.1109/CEC.2013.6557743
Alan Díaz-Manríquez, G. T. Pulido, C. Coello, R. Becerra
In recent years, the development of selection mechanisms based on performance indicators has become an important trend in algorithmic design. Hereof, the hypervolume has been the most popular choice. Multi-objective evolutionary algorithms (MOEAs) based on this indicator seem to be a good choice for dealing with many-objective optimization problems. However, their main drawback is that such algorithms are typically computationally expensive. This has motivated some recent research in which the use of other performance indicators has been explored. Here, we propose an efficient mechanism to integrate the R2 indicator to a modified version of Goldberg's nondominated sorting method, in order to rank the individuals of a MOEA. Our proposed ranking scheme is coupled to two different search engines, resulting in two new MOEAs. These MOEAs are validated using several test problems and performance measures commonly adopted in the specialized literature. Results indicate that the proposed ranking approach gives rise to effective MOEAs, which produce results that are competitive with respect to those obtained by three well-known MOEAs. Additionally, we validate our resulting MOEAs in many-objective optimization problems, in which our proposed ranking scheme shows its main advantage, since it is able to outperform a hypervolume-based MOEA, requiring a much lower computational time.
{"title":"A ranking method based on the R2 indicator for many-objective optimization","authors":"Alan Díaz-Manríquez, G. T. Pulido, C. Coello, R. Becerra","doi":"10.1109/CEC.2013.6557743","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557743","url":null,"abstract":"In recent years, the development of selection mechanisms based on performance indicators has become an important trend in algorithmic design. Hereof, the hypervolume has been the most popular choice. Multi-objective evolutionary algorithms (MOEAs) based on this indicator seem to be a good choice for dealing with many-objective optimization problems. However, their main drawback is that such algorithms are typically computationally expensive. This has motivated some recent research in which the use of other performance indicators has been explored. Here, we propose an efficient mechanism to integrate the R2 indicator to a modified version of Goldberg's nondominated sorting method, in order to rank the individuals of a MOEA. Our proposed ranking scheme is coupled to two different search engines, resulting in two new MOEAs. These MOEAs are validated using several test problems and performance measures commonly adopted in the specialized literature. Results indicate that the proposed ranking approach gives rise to effective MOEAs, which produce results that are competitive with respect to those obtained by three well-known MOEAs. Additionally, we validate our resulting MOEAs in many-objective optimization problems, in which our proposed ranking scheme shows its main advantage, since it is able to outperform a hypervolume-based MOEA, requiring a much lower computational time.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123686393","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 : 2013-06-20DOI: 10.1109/CEC.2013.6557661
S. Baluja, Michele Covell
In the context of stochastic search, once regions of high performance are found, having the property that small changes in the candidate solution correspond to searching nearby neighborhoods provides the ability to perform effective local optimization. To achieve this, Gray Codes are often employed for encoding ordinal points or discretized real numbers. In this paper, we present a method to label similar and/or close points within arbitrary graphs with small Hamming distances. The resultant point labels can be viewed as an approximate high-dimensional variant of Gray Codes. The labeling procedure is useful for any task in which the solution requires the search algorithm to select a small subset of items out of many. A large number of empirical results using these encodings with a combination of genetic algorithms and hill-climbing are presented.
{"title":"Point representation for local optimization","authors":"S. Baluja, Michele Covell","doi":"10.1109/CEC.2013.6557661","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557661","url":null,"abstract":"In the context of stochastic search, once regions of high performance are found, having the property that small changes in the candidate solution correspond to searching nearby neighborhoods provides the ability to perform effective local optimization. To achieve this, Gray Codes are often employed for encoding ordinal points or discretized real numbers. In this paper, we present a method to label similar and/or close points within arbitrary graphs with small Hamming distances. The resultant point labels can be viewed as an approximate high-dimensional variant of Gray Codes. The labeling procedure is useful for any task in which the solution requires the search algorithm to select a small subset of items out of many. A large number of empirical results using these encodings with a combination of genetic algorithms and hill-climbing are presented.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114238074","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 : 2013-06-20DOI: 10.1109/CEC.2013.6557950
A. Aleti, I. Moser
The performance of an Evolutionary Algorithm (EA) is greatly affected by the settings of its strategy parameters. An effective solution to the parameterisation problem is adaptive parameter control, which applies learning methods that use feedback from the optimisation process to evaluate the effect of parameter value choices and adjust the parameter values over the iterations. At every iteration of an EA, the performance of an EA is reported and employed by the feedback mechanism as an indication of the success of the parameterisation of the algorithm instance. Many approaches to collect information about the algorithm's performance exist in single objective optimisation. In this work, we review the most recent and prominent approaches. In multiobjective optimisation, establishing a single scalar which can report the algorithm's performance as feedback for adaptive parameter control is a complex task. Existing performance measures of multiobjective optimisation are generally used as feedback for the optimisation process. We discuss the properties of these measures and present an empirical evaluation of the binary hypervolume and ϵ+-indicators as feedback for adaptive parameter control.
{"title":"Studying feedback mechanisms for adaptive parameter control in evolutionary algorithms","authors":"A. Aleti, I. Moser","doi":"10.1109/CEC.2013.6557950","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557950","url":null,"abstract":"The performance of an Evolutionary Algorithm (EA) is greatly affected by the settings of its strategy parameters. An effective solution to the parameterisation problem is adaptive parameter control, which applies learning methods that use feedback from the optimisation process to evaluate the effect of parameter value choices and adjust the parameter values over the iterations. At every iteration of an EA, the performance of an EA is reported and employed by the feedback mechanism as an indication of the success of the parameterisation of the algorithm instance. Many approaches to collect information about the algorithm's performance exist in single objective optimisation. In this work, we review the most recent and prominent approaches. In multiobjective optimisation, establishing a single scalar which can report the algorithm's performance as feedback for adaptive parameter control is a complex task. Existing performance measures of multiobjective optimisation are generally used as feedback for the optimisation process. We discuss the properties of these measures and present an empirical evaluation of the binary hypervolume and ϵ+-indicators as feedback for adaptive parameter control.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114335356","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 : 2013-06-20DOI: 10.1109/CEC.2013.6557587
D. Ashlock, Carolyn Pugh
Cellular automata are known to be capable of Turing-complete computation and yet “programming” them to do particular tasks can be quite daunting. In this paper we use single parent crossover as a means of transferring information between successive evolving populations to create rules for cellular automata that have proscribed shapes. The proscription of regions where the automata are permitted to grow is the reason they are called bonsai automata. This work follows earlier work on apoptotic cellular automata that simply exhibit self-limited growth. The correct choice of single parents permits enormous improvement in the performance of evolutionary algorithms searching for automata that satisfy particular bonsai templates. In this study, we demonstrate that single parent techniques make meeting shape constraints on the growth of CAs possible at all in some cases. This study also introduces range niche specialization to control problems with the cloning of ancestors used for single parent crossover in an evolving population. This study demonstrates that different bonsai shapes have highly variable difficulty. It is also shown that automata evolved to satisfy one bonsai template may be needed to enable, via single parent crossover, solutions for another template. The use of bonsai techniques yields many automata not found during studies of apoptotic automata demonstrating that the technique encourages exploration of different parts of the fitness landscape.
{"title":"Evolutionary cellular automata bonsai","authors":"D. Ashlock, Carolyn Pugh","doi":"10.1109/CEC.2013.6557587","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557587","url":null,"abstract":"Cellular automata are known to be capable of Turing-complete computation and yet “programming” them to do particular tasks can be quite daunting. In this paper we use single parent crossover as a means of transferring information between successive evolving populations to create rules for cellular automata that have proscribed shapes. The proscription of regions where the automata are permitted to grow is the reason they are called bonsai automata. This work follows earlier work on apoptotic cellular automata that simply exhibit self-limited growth. The correct choice of single parents permits enormous improvement in the performance of evolutionary algorithms searching for automata that satisfy particular bonsai templates. In this study, we demonstrate that single parent techniques make meeting shape constraints on the growth of CAs possible at all in some cases. This study also introduces range niche specialization to control problems with the cloning of ancestors used for single parent crossover in an evolving population. This study demonstrates that different bonsai shapes have highly variable difficulty. It is also shown that automata evolved to satisfy one bonsai template may be needed to enable, via single parent crossover, solutions for another template. The use of bonsai techniques yields many automata not found during studies of apoptotic automata demonstrating that the technique encourages exploration of different parts of the fitness landscape.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114756149","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 : 2013-06-20DOI: 10.1109/CEC.2013.6557809
Samadhi Nallaperuma, Andrew M. Sutton, F. Neumann
Recently, Sutton and Neumann [1] have studied evolutionary algorithms for the Euclidean traveling salesman problem by parameterized runtime analyses taking into account the number of inner points k and the number of cities n. They have shown that simple evolutionary algorithms are XP-algorithms for the problem, i.e., they obtain an optimal solution in expected time O(ng(k)) where g(k) is a function only depending on k. We extend these investigations and design two evolutionary algorithms for the Euclidean Traveling Salesperson problem that run in expected time g(k) · poly(n) where k is a parameter denoting the number inner points for the given TSP instance, i.e., they are fixed-parameter tractable evolutionary algorithms for the Euclidean TSP parameterized by the number of inner points. While our first approach is mainly of theoretical interest, our second approach leverages problem structure by directly searching for good orderings of the inner points and provides a novel and highly effective way of tackling this important problem. Our experimental results show that searching for a permutation on the inner points is a significantly powerful practical strategy.
{"title":"Fixed-parameter evolutionary algorithms for the Euclidean Traveling Salesperson problem","authors":"Samadhi Nallaperuma, Andrew M. Sutton, F. Neumann","doi":"10.1109/CEC.2013.6557809","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557809","url":null,"abstract":"Recently, Sutton and Neumann [1] have studied evolutionary algorithms for the Euclidean traveling salesman problem by parameterized runtime analyses taking into account the number of inner points k and the number of cities n. They have shown that simple evolutionary algorithms are XP-algorithms for the problem, i.e., they obtain an optimal solution in expected time O(ng(k)) where g(k) is a function only depending on k. We extend these investigations and design two evolutionary algorithms for the Euclidean Traveling Salesperson problem that run in expected time g(k) · poly(n) where k is a parameter denoting the number inner points for the given TSP instance, i.e., they are fixed-parameter tractable evolutionary algorithms for the Euclidean TSP parameterized by the number of inner points. While our first approach is mainly of theoretical interest, our second approach leverages problem structure by directly searching for good orderings of the inner points and provides a novel and highly effective way of tackling this important problem. Our experimental results show that searching for a permutation on the inner points is a significantly powerful practical strategy.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114787174","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 : 2013-06-20DOI: 10.1109/CEC.2013.6557970
A. Banerjee
The problem of identifying core patterns with the correct neighborhood parameters is a major challenge for density-based clustering techniques derived from the popular DBSCAN algorithm. An evolutionary approach to optimizing the assignment of core patterns is proposed in this paper. Key ideas presented here include a genetic representation that associates distinct neighborhood parameters with potential core patterns and specialized crossover and mutation operators. The evolutionary framework is based on the multi-objective NSGA-II algorithm, with simplified fitness measures derived from local (neighborhood) information. Clustering experiments on both synthetic and benchmark clustering datasets are presented and results are compared to the original DBSCAN, fuzzy DBSCAN and k-means.
{"title":"On evolving neighborhood parameters for fuzzy density clustering","authors":"A. Banerjee","doi":"10.1109/CEC.2013.6557970","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557970","url":null,"abstract":"The problem of identifying core patterns with the correct neighborhood parameters is a major challenge for density-based clustering techniques derived from the popular DBSCAN algorithm. An evolutionary approach to optimizing the assignment of core patterns is proposed in this paper. Key ideas presented here include a genetic representation that associates distinct neighborhood parameters with potential core patterns and specialized crossover and mutation operators. The evolutionary framework is based on the multi-objective NSGA-II algorithm, with simplified fitness measures derived from local (neighborhood) information. Clustering experiments on both synthetic and benchmark clustering datasets are presented and results are compared to the original DBSCAN, fuzzy DBSCAN and k-means.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124074355","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 : 2013-06-20DOI: 10.1109/CEC.2013.6557660
K. J. Gurubel, E. Sánchez, S. Carlos-Hernandez, Fernando Ornelas
This paper proposes a hybrid intelligent inverse optimal control for trajectory tracking based on a neural observer and a fuzzy supervisor for an anaerobic digestion process, in order to maximize methane production. A nonlinear discrete-time recurrent high order neural observer (RHONO) is used to estimate biomass concentration and substrate degradation in a continuous stirred tank reactor. The control law calculates dilution rate and bicarbonate supply, and a Takagi-Sugeno supervisor based on the estimation of biomass, selects and applies the most adequate control action, allowing a smooth switching between open loop and closed loop. A Particle Swarm Optimization (PSO) algorithm is employed to determine the matrix P for inverse optimal control in order to improve tracking results. The applicability of the proposed scheme is illustrated via simulations.
{"title":"PSO hybrid intelligent inverse optimal control for an anaerobic process","authors":"K. J. Gurubel, E. Sánchez, S. Carlos-Hernandez, Fernando Ornelas","doi":"10.1109/CEC.2013.6557660","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557660","url":null,"abstract":"This paper proposes a hybrid intelligent inverse optimal control for trajectory tracking based on a neural observer and a fuzzy supervisor for an anaerobic digestion process, in order to maximize methane production. A nonlinear discrete-time recurrent high order neural observer (RHONO) is used to estimate biomass concentration and substrate degradation in a continuous stirred tank reactor. The control law calculates dilution rate and bicarbonate supply, and a Takagi-Sugeno supervisor based on the estimation of biomass, selects and applies the most adequate control action, allowing a smooth switching between open loop and closed loop. A Particle Swarm Optimization (PSO) algorithm is employed to determine the matrix P for inverse optimal control in order to improve tracking results. The applicability of the proposed scheme is illustrated via simulations.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127606161","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 : 2013-06-20DOI: 10.1109/CEC.2013.6557560
Corrado Grappiolo, J. Togelius, Georgios N. Yannakakis
We present a new evolutionary algorithm for community structure detection in both undirected and unweighted (sparse) graphs and fully connected weighted digraphs (complete networks). Previous investigations have found that, although evolutionary computation can identify community structure in complete networks, this approach seems to scale badly due to solutions with the wrong number of communities dominating the population. The new algorithm is based on a niching model, where separate compartments of the population contain candidate solutions with different numbers of communities. We experimentally compare the new algorithm to the well-known algorithms of Pizzuti and Tasgin, and find that we outperform those algorithms for sparse graphs under some conditions, and drastically outperform them on complete networks under all tested conditions.
{"title":"Shifting niches for community structure detection","authors":"Corrado Grappiolo, J. Togelius, Georgios N. Yannakakis","doi":"10.1109/CEC.2013.6557560","DOIUrl":"https://doi.org/10.1109/CEC.2013.6557560","url":null,"abstract":"We present a new evolutionary algorithm for community structure detection in both undirected and unweighted (sparse) graphs and fully connected weighted digraphs (complete networks). Previous investigations have found that, although evolutionary computation can identify community structure in complete networks, this approach seems to scale badly due to solutions with the wrong number of communities dominating the population. The new algorithm is based on a niching model, where separate compartments of the population contain candidate solutions with different numbers of communities. We experimentally compare the new algorithm to the well-known algorithms of Pizzuti and Tasgin, and find that we outperform those algorithms for sparse graphs under some conditions, and drastically outperform them on complete networks under all tested conditions.","PeriodicalId":211988,"journal":{"name":"2013 IEEE Congress on Evolutionary Computation","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127887163","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：