Pub Date : 2003-12-08DOI: 10.1109/CEC.2003.1299412
Kazuyuki Ito, A. Gofuku
Acquiring adaptive behaviors of robots automatically is one of the most interesting topics of the evolutionary systems. In previous works, we have developed an adaptive autonomous control method for redundant robots. The QDSEGA is one of the methods that we have proposed for them. The QDSEGA is realized by combining Q-learning and GA, and it can acquire suitable behaviors by adapting a movement of a robot for a task. In this paper, we focus on the adaptability of the QDSEGA and discuss the robustness of the autonomous redundant robot that is controlled by the QDSEGA. To demonstrate the effectiveness of the QDSEGA, simulations of obstacle avoidance by a 10-link manipulator in the changeable environment and locomotion by a 12-legged robot with failures have been carried out, and as a result, adaptive behaviors for each environment and each broken body have emerged.
{"title":"Emergence of adaptive behaviors by redundant robots - robustness to changes environment and failures","authors":"Kazuyuki Ito, A. Gofuku","doi":"10.1109/CEC.2003.1299412","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299412","url":null,"abstract":"Acquiring adaptive behaviors of robots automatically is one of the most interesting topics of the evolutionary systems. In previous works, we have developed an adaptive autonomous control method for redundant robots. The QDSEGA is one of the methods that we have proposed for them. The QDSEGA is realized by combining Q-learning and GA, and it can acquire suitable behaviors by adapting a movement of a robot for a task. In this paper, we focus on the adaptability of the QDSEGA and discuss the robustness of the autonomous redundant robot that is controlled by the QDSEGA. To demonstrate the effectiveness of the QDSEGA, simulations of obstacle avoidance by a 10-link manipulator in the changeable environment and locomotion by a 12-legged robot with failures have been carried out, and as a result, adaptive behaviors for each environment and each broken body have emerged.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116338297","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299771
J. Heinonen, F. Pettersson
A genetic algorithm is used to calculate production schedules for a specific type of batch-mode manufacturing process. Previous scheduling efforts included both time discretised MILP as well as continuous-time MILP, both of which were outperformed by the GA when measured in calculation times and final schedule accuracy. The approach is two-folded and an algorithm to reproduce it on similar processes is presented.
{"title":"Scheduling a specific type of batch process with evolutionary computation","authors":"J. Heinonen, F. Pettersson","doi":"10.1109/CEC.2003.1299771","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299771","url":null,"abstract":"A genetic algorithm is used to calculate production schedules for a specific type of batch-mode manufacturing process. Previous scheduling efforts included both time discretised MILP as well as continuous-time MILP, both of which were outperformed by the GA when measured in calculation times and final schedule accuracy. The approach is two-folded and an algorithm to reproduce it on similar processes is presented.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121507198","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299420
A. M. Abdelbar, M. Mokhtar
Abduction is the process of proceeding from data describing a set of observations or events, to a set of hypotheses which best explains or accounts for the data. Cost-based abduction (CBA) is a formalism in which evidence to be explained is treated as a goal to be proven, proofs have costs based on how much needs to be assumed to complete the proof, and the set of assumptions needed to complete the least-cost proof are taken as the best explanation for the given evidence. We apply a k-elitist variation on the max-min ant system (MMAS) to CBA, in which the k-best ants are allowed to update the global pheromone trace array in every iteration; in the original MMAS, only the single best ant updates the trace array (thus, it can be considered 1-elitist). Applying our technique to several large CBA instances, we find that our k-elitist approach, with k varying in our experiments from 1 to 15, returns lower-cost proofs on average than the original MMAS. A test of statistical significance is used to verify that the differences in performance are statistically significant.
{"title":"A k-elitist max-min ant system approach to cost-based abduction","authors":"A. M. Abdelbar, M. Mokhtar","doi":"10.1109/CEC.2003.1299420","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299420","url":null,"abstract":"Abduction is the process of proceeding from data describing a set of observations or events, to a set of hypotheses which best explains or accounts for the data. Cost-based abduction (CBA) is a formalism in which evidence to be explained is treated as a goal to be proven, proofs have costs based on how much needs to be assumed to complete the proof, and the set of assumptions needed to complete the least-cost proof are taken as the best explanation for the given evidence. We apply a k-elitist variation on the max-min ant system (MMAS) to CBA, in which the k-best ants are allowed to update the global pheromone trace array in every iteration; in the original MMAS, only the single best ant updates the trace array (thus, it can be considered 1-elitist). Applying our technique to several large CBA instances, we find that our k-elitist approach, with k varying in our experiments from 1 to 15, returns lower-cost proofs on average than the original MMAS. A test of statistical significance is used to verify that the differences in performance are statistically significant.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124465181","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299921
T.T.H. Luong, Q.T. Pham
All test problems in the optimization and genetic algorithm (GA) literature involve analytical objective functions, which can be calculated exactly (to within floating point accuracy) using elementary operations and functions. However, almost al practical chemical engineering optimization problems involve sets of nonlinear equations or ordinary or partial differential equations that must be solved by some numerical methods (iterative root finding, finite differences, Rung Kutta, etc.) which inherent rounding and truncation errors. It is suspected that evolutionary methods such as genetic algorithms are better than classical deterministic methods for these problems. This paper aims to test this hypothesis by comparing the performance of two classical deterministic methods and a GA method on some representative engineering problems.
{"title":"A comparison of the performance of classical methods and genetic algorithms for optimization problems involving numerical models","authors":"T.T.H. Luong, Q.T. Pham","doi":"10.1109/CEC.2003.1299921","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299921","url":null,"abstract":"All test problems in the optimization and genetic algorithm (GA) literature involve analytical objective functions, which can be calculated exactly (to within floating point accuracy) using elementary operations and functions. However, almost al practical chemical engineering optimization problems involve sets of nonlinear equations or ordinary or partial differential equations that must be solved by some numerical methods (iterative root finding, finite differences, Rung Kutta, etc.) which inherent rounding and truncation errors. It is suspected that evolutionary methods such as genetic algorithms are better than classical deterministic methods for these problems. This paper aims to test this hypothesis by comparing the performance of two classical deterministic methods and a GA method on some representative engineering problems.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"142 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124512001","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299864
Mark R. Knarr, M. Goltz, G. Lamont, Junqi Huang
Combining horizontal flow treatment wells (HFTWs) with in situ biodegradation is an innovative approach with the potential to remediate perchlorate-contaminated groundwater. A model has been developed that combines the groundwater flow induced by HFTWs with biodegradation processes that result from using the HFTWs to mix electron donor into perchlorate-contaminated groundwater. The model can be used to select engineering design parameters that optimize performance under given site conditions. In particular, one desires to design a system that 1) maximizes perchlorate destruction, 2) minimizes treatment expense, and 3) attains regulatory limits on downgradient contaminant concentrations. Unfortunately, for a relatively complex technology like in situ bioremediation, system optimization is not straightforward. In this study, a general multi-objective parallel evolutionary algorithm call GENMOP is developed and used to stochastically determine design parameter values (flow rate, well spacing, concentration of injected electron donor, and injection schedule) in order to maximize perchlorate destruction while minimizing cost. Results indicate that the relationship between perchlorate mass removal and operating cost is positively correlated and nonlinear. For equivalent operating times and costs, the solutions show that the technology achieves higher perchlorate mass removals for a site having both higher hydraulic conductivity as well as higher initial perchlorate concentrations.
{"title":"In situ bioremediation of perchlorate-contaminated groundwater using a multi-objective parallel evolutionary algorithm","authors":"Mark R. Knarr, M. Goltz, G. Lamont, Junqi Huang","doi":"10.1109/CEC.2003.1299864","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299864","url":null,"abstract":"Combining horizontal flow treatment wells (HFTWs) with in situ biodegradation is an innovative approach with the potential to remediate perchlorate-contaminated groundwater. A model has been developed that combines the groundwater flow induced by HFTWs with biodegradation processes that result from using the HFTWs to mix electron donor into perchlorate-contaminated groundwater. The model can be used to select engineering design parameters that optimize performance under given site conditions. In particular, one desires to design a system that 1) maximizes perchlorate destruction, 2) minimizes treatment expense, and 3) attains regulatory limits on downgradient contaminant concentrations. Unfortunately, for a relatively complex technology like in situ bioremediation, system optimization is not straightforward. In this study, a general multi-objective parallel evolutionary algorithm call GENMOP is developed and used to stochastically determine design parameter values (flow rate, well spacing, concentration of injected electron donor, and injection schedule) in order to maximize perchlorate destruction while minimizing cost. Results indicate that the relationship between perchlorate mass removal and operating cost is positively correlated and nonlinear. For equivalent operating times and costs, the solutions show that the technology achieves higher perchlorate mass removals for a site having both higher hydraulic conductivity as well as higher initial perchlorate concentrations.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127618562","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299947
F. Corno, E. Sánchez, Giovanni Squillero
In this paper, Core War, a very peculiar game popular in mid 80's, is exploited as a benchmark to improve the /spl mu/GP, an evolutionary algorithm able to generate touring-complete, realistic assembly programs. Two techniques were analyzed: coevolution and a modified island model. Experimental results showed that the former is essential in the beginning of the evolutionary process, but may be deceptive in the end. Differently, the latter enables focusing the search on specific region of the search space and lead to dramatic improvements. The use of both techniques to help the /spl mu/GP in its real task (test program generation for microprocessor) is currently being evaluated.
{"title":"Exploiting co-evolution and a modified island model to climb the Core War hill","authors":"F. Corno, E. Sánchez, Giovanni Squillero","doi":"10.1109/CEC.2003.1299947","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299947","url":null,"abstract":"In this paper, Core War, a very peculiar game popular in mid 80's, is exploited as a benchmark to improve the /spl mu/GP, an evolutionary algorithm able to generate touring-complete, realistic assembly programs. Two techniques were analyzed: coevolution and a modified island model. Experimental results showed that the former is essential in the beginning of the evolutionary process, but may be deceptive in the end. Differently, the latter enables focusing the search on specific region of the search space and lead to dramatic improvements. The use of both techniques to help the /spl mu/GP in its real task (test program generation for microprocessor) is currently being evaluated.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127730027","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299745
Stefan Janson, M. Middendorf
A hierarchical version of the particle swarm optimization method called H-PSO is introduced. In H-PSO the particles are arranged in a dynamic hierarchy that is used to define a neighborhood structure. Depending on the quality of their so far best found solution the particles move up or down the hierarchy so that good particles have a higher influence on the swarm. Moreover, the hierarchy is used to define different search properties for the particles. Several variants of H-PSO are compared experimentally with variants of the standard PSO.
{"title":"A hierarchical particle swarm optimizer","authors":"Stefan Janson, M. Middendorf","doi":"10.1109/CEC.2003.1299745","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299745","url":null,"abstract":"A hierarchical version of the particle swarm optimization method called H-PSO is introduced. In H-PSO the particles are arranged in a dynamic hierarchy that is used to define a neighborhood structure. Depending on the quality of their so far best found solution the particles move up or down the hierarchy so that good particles have a higher influence on the swarm. Moreover, the hierarchy is used to define different search properties for the particles. Several variants of H-PSO are compared experimentally with variants of the standard PSO.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126435707","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299799
A. Auyeung, A. Abraham
Sorting by reversals is an important problem in inferring the evolutionary relationship between two genomes. The problem of sorting unsigned permutation has been proven to be NP-hard. The best guaranteed error bounded is the 3/2-approximation algorithm. However, the problem of sorting signed permutation can be solved easily. Fast algorithms have been developed both for finding the sorting sequence and finding the reversal distance of signed permutation. We present a way to view the problem of sorting unsigned permutation as signed permutation. And the problem can then be seen as searching an optimal signed permutation in all 2/sup n/ corresponding signed permutations. We use genetic algorithm to conduct the search. Our experimental result shows that the proposed method outperform the 3/2-approximation algorithm.
{"title":"Estimating genome reversal distance by genetic algorithm","authors":"A. Auyeung, A. Abraham","doi":"10.1109/CEC.2003.1299799","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299799","url":null,"abstract":"Sorting by reversals is an important problem in inferring the evolutionary relationship between two genomes. The problem of sorting unsigned permutation has been proven to be NP-hard. The best guaranteed error bounded is the 3/2-approximation algorithm. However, the problem of sorting signed permutation can be solved easily. Fast algorithms have been developed both for finding the sorting sequence and finding the reversal distance of signed permutation. We present a way to view the problem of sorting unsigned permutation as signed permutation. And the problem can then be seen as searching an optimal signed permutation in all 2/sup n/ corresponding signed permutations. We use genetic algorithm to conduct the search. Our experimental result shows that the proposed method outperform the 3/2-approximation algorithm.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126471565","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299776
W. A. Greene
We present a clustering algorithm which is unsupervised, incremental, and hierarchical. The algorithm is distance-based and creates centroids. Then we combine the power of evolutionary forces with the clustering algorithm, counting on good clusterings to evolve to yet better ones. We apply our approach to standard data sets, and get very good results. Finally, we use bagging to pool the results of different clustering trials, and again get very good results.
{"title":"Unsupervised hierarchical clustering via a genetic algorithm","authors":"W. A. Greene","doi":"10.1109/CEC.2003.1299776","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299776","url":null,"abstract":"We present a clustering algorithm which is unsupervised, incremental, and hierarchical. The algorithm is distance-based and creates centroids. Then we combine the power of evolutionary forces with the clustering algorithm, counting on good clusterings to evolve to yet better ones. We apply our approach to standard data sets, and get very good results. Finally, we use bagging to pool the results of different clustering trials, and again get very good results.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128096091","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299588
Kit Yan Chan, Mehmet Emin Aydin, T. Fogarty
Epistasis is a measure of interdependence between genes and an indicator of problem difficulty in genetic algorithms. Many researches have concentrated on the epistasis measure in binary coded representation in genetic algorithms. However, a few attempts for epistasis measure in real-coded representation have been reported in the literature. In this paper, we have demonstrated how to use the approach of analysis of variance (ANOVA) to estimate the epistasis in real-coded representation. The approach is useful to analyse epistasis in genetic algorithms in a more detailed level. Examples have been given for showing how to use ANOVA for measuring the amount of epistasis in parametrical problems, and then we have applied this epistatic information provided by ANOVA to improve the performance of genetic algorithm.
{"title":"An epistasis measure based on the analysis of variance for the real-coded representation in genetic algorithms","authors":"Kit Yan Chan, Mehmet Emin Aydin, T. Fogarty","doi":"10.1109/CEC.2003.1299588","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299588","url":null,"abstract":"Epistasis is a measure of interdependence between genes and an indicator of problem difficulty in genetic algorithms. Many researches have concentrated on the epistasis measure in binary coded representation in genetic algorithms. However, a few attempts for epistasis measure in real-coded representation have been reported in the literature. In this paper, we have demonstrated how to use the approach of analysis of variance (ANOVA) to estimate the epistasis in real-coded representation. The approach is useful to analyse epistasis in genetic algorithms in a more detailed level. Examples have been given for showing how to use ANOVA for measuring the amount of epistasis in parametrical problems, and then we have applied this epistatic information provided by ANOVA to improve the performance of genetic algorithm.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121734854","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：