Slideshows by User: pelikan http://www.slideshare.net/ http://www.slideshare.net/images/logo.gif Slideshows by User: pelikan http://www.slideshare.net/ Wed, 22 Jul 2009 03:23:10 GMT SlideShare feed for Slideshows by User: pelikan Effects of a Deterministic Hill climber on hBOA http://www.slideshare.net/pelikan/effects-of-a-deterministic-hill-climber-on-hboa
Hybridization of global and local search algorithms is a well-established technique for enhancing the efficiency of search algorithms. Hybridizing estimation of distribution algorithms (EDAs) has been repeatedly shown to produce better performance than either the global or local search algorithm alone. The hierarchical Bayesian optimization algorithm (hBOA) is an advanced EDA which has previously been shown to benefit from hybridization with a local searcher. This paper examines the effects of combining hBOA with a deterministic hill climber (DHC). Experiments reveal that allowing DHC to find the local optima makes model building and decision making much easier for hBOA. This reduces the minimum population size required to find the global optimum, which substantially improves overall performance.]]>
Hybridization of global and local search algorithms is a well-established technique for enhancing the efficiency of search algorithms. Hybridizing estimation of distribution algorithms (EDAs) has been repeatedly shown to produce better performance than either the global or local search algorithm alone. The hierarchical Bayesian optimization algorithm (hBOA) is an advanced EDA which has previously been shown to benefit from hybridization with a local searcher. This paper examines the effects of combining hBOA with a deterministic hill climber (DHC). Experiments reveal that allowing DHC to find the local optima makes model building and decision making much easier for hBOA. This reduces the minimum population size required to find the global optimum, which substantially improves overall performance.]]> Wed, 22 Jul 2009 03:23:10 GMT http://www.slideshare.net/pelikan/effects-of-a-deterministic-hill-climber-on-hboa pelikan@slideshare.net(pelikan) Effects of a Deterministic Hill climber on hBOA pelikan Hybridization of global and local search algorithms is a well-established technique for enhancing the efficiency of search algorithms. Hybridizing estimation of distribution algorithms (EDAs) has been repeatedly shown to produce better performance than either the global or local search algorithm alone. The hierarchical Bayesian optimization algorithm (hBOA) is an advanced EDA which has previously been shown to benefit from hybridization with a local searcher. This paper examines the effects of combining hBOA with a deterministic hill climber (DHC). Experiments reveal that allowing DHC to find the local optima makes model building and decision making much easier for hBOA. This reduces the minimum population size required to find the global optimum, which substantially improves overall performance. <img src="http://cdn.slidesharecdn.com/dhc-hboa-090721222318-phpapp02-thumbnail-2?1248233013" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> Hybridization of global and local search algorithms is a well-established technique for enhancing the efficiency of search algorithms. Hybridizing estimation of distribution algorithms (EDAs) has been repeatedly shown to produce better performance than either the global or local search algorithm alone. The hierarchical Bayesian optimization algorithm (hBOA) is an advanced EDA which has previously been shown to benefit from hybridization with a local searcher. This paper examines the effects of combining hBOA with a deterministic hill climber (DHC). Experiments reveal that allowing DHC to find the local optima makes model building and decision making much easier for hBOA. This reduces the minimum population size required to find the global optimum, which substantially improves overall performance. Effects of a Deterministic Hill climber on hBOA
View more documents from pelikan.
]]> 145 0 http://cdn.slidesharecdn.com/dhc-hboa-090721222318-phpapp02-thumbnail-2?1248233013 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Intelligent Bias of Network Structures in the Hierarchical BOA http://www.slideshare.net/pelikan/ntelligent-bias-of-network-structures-in-the-hierarchical-boa
One of the primary advantages of estimation of distribution algorithms (EDAs) over many other stochastic optimization techniques is that they supply us with a roadmap of how they solve a problem. This roadmap consists of a sequence of probabilistic models of candidate solutions of increasing quality. The first model in this sequence would typically encode the uniform distribution over all admissible solutions whereas the last model would encode a distribution that generates at least one global optimum with high probability. It has been argued that exploiting this knowledge should improve EDA performance when solving similar problems. This paper presents an approach to bias the building of Bayesian network models in the hierarchical Bayesian optimization algorithm (hBOA) using information gathered from models generated during previous hBOA runs on similar problems. The approach is evaluated on trap-5 and 2D spin glass problems.]]>
One of the primary advantages of estimation of distribution algorithms (EDAs) over many other stochastic optimization techniques is that they supply us with a roadmap of how they solve a problem. This roadmap consists of a sequence of probabilistic models of candidate solutions of increasing quality. The first model in this sequence would typically encode the uniform distribution over all admissible solutions whereas the last model would encode a distribution that generates at least one global optimum with high probability. It has been argued that exploiting this knowledge should improve EDA performance when solving similar problems. This paper presents an approach to bias the building of Bayesian network models in the hierarchical Bayesian optimization algorithm (hBOA) using information gathered from models generated during previous hBOA runs on similar problems. The approach is evaluated on trap-5 and 2D spin glass problems.]]> Wed, 22 Jul 2009 03:21:50 GMT http://www.slideshare.net/pelikan/ntelligent-bias-of-network-structures-in-the-hierarchical-boa pelikan@slideshare.net(pelikan) Intelligent Bias of Network Structures in the Hierarchical BOA pelikan One of the primary advantages of estimation of distribution algorithms (EDAs) over many other stochastic optimization techniques is that they supply us with a roadmap of how they solve a problem. This roadmap consists of a sequence of probabilistic models of candidate solutions of increasing quality. The first model in this sequence would typically encode the uniform distribution over all admissible solutions whereas the last model would encode a distribution that generates at least one global optimum with high probability. It has been argued that exploiting this knowledge should improve EDA performance when solving similar problems. This paper presents an approach to bias the building of Bayesian network models in the hierarchical Bayesian optimization algorithm (hBOA) using information gathered from models generated during previous hBOA runs on similar problems. The approach is evaluated on trap-5 and 2D spin glass problems. <img src="http://cdn.slidesharecdn.com/intelligent-bias-090721222200-phpapp01-thumbnail-2?1248233230" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> One of the primary advantages of estimation of distribution algorithms (EDAs) over many other stochastic optimization techniques is that they supply us with a roadmap of how they solve a problem. This roadmap consists of a sequence of probabilistic models of candidate solutions of increasing quality. The first model in this sequence would typically encode the uniform distribution over all admissible solutions whereas the last model would encode a distribution that generates at least one global optimum with high probability. It has been argued that exploiting this knowledge should improve EDA performance when solving similar problems. This paper presents an approach to bias the building of Bayesian network models in the hierarchical Bayesian optimization algorithm (hBOA) using information gathered from models generated during previous hBOA runs on similar problems. The approach is evaluated on trap-5 and 2D spin glass problems. Intelligent Bias of Network Structures in the Hierarchical BOA
View more documents from pelikan.
]]> 193 0 http://cdn.slidesharecdn.com/intelligent-bias-090721222200-phpapp01-thumbnail-2?1248233230 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Analysis of Evolutionary Algorithms on the One-Dimensional Spin Glass with Power-Law Interactions http://www.slideshare.net/pelikan/analysis-of-evolutionary-algorithms-on-the-onedimensional-spin-glass-with-powerlaw-interactions
]]>
]]> Mon, 13 Jul 2009 15:13:32 GMT http://www.slideshare.net/pelikan/analysis-of-evolutionary-algorithms-on-the-onedimensional-spin-glass-with-powerlaw-interactions pelikan@slideshare.net(pelikan) Analysis of Evolutionary Algorithms on the One-Dimensional Spin Glass with Power-Law Interactions pelikan <img src="http://cdn.slidesharecdn.com/sk-ring-090713101332-phpapp01-thumbnail-2?1247498036" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> Analysis of Evolutionary Algorithms on the One-Dimensional Spin Glass with Power-Law Interactions
View more documents from pelikan.
]]> 152 0 http://cdn.slidesharecdn.com/sk-ring-090713101332-phpapp01-thumbnail-2?1247498036 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Performance of Evolutionary Algorithms on NK Landscapes with Nearest Neighbor Interactions and Tunable Overlap http://www.slideshare.net/pelikan/performance-of-evolutionary-algorithms-on-nk-landscapes-with-nearest-neighbor-interactions-and-tunable-overlap
]]>
]]> Mon, 13 Jul 2009 15:13:30 GMT http://www.slideshare.net/pelikan/performance-of-evolutionary-algorithms-on-nk-landscapes-with-nearest-neighbor-interactions-and-tunable-overlap pelikan@slideshare.net(pelikan) Performance of Evolutionary Algorithms on NK Landscapes with Nearest Neighbor Interactions and Tunable Overlap pelikan <img src="http://cdn.slidesharecdn.com/nk-nearest-neighbor-090713101335-phpapp02-thumbnail-2?1247498062" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> Performance of Evolutionary Algorithms on NK Landscapes with Nearest Neighbor Interactions and Tunable Overlap
View more documents from pelikan.
]]> 277 0 http://cdn.slidesharecdn.com/nk-nearest-neighbor-090713101335-phpapp02-thumbnail-2?1247498062 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Initial-Population Bias in the Univariate Estimation of Distribution Algorithm http://www.slideshare.net/pelikan/initialpopulation-bias-in-the-univariate-estimation-of-distribution-algorithm
]]>
]]> Mon, 13 Jul 2009 06:46:39 GMT http://www.slideshare.net/pelikan/initialpopulation-bias-in-the-univariate-estimation-of-distribution-algorithm pelikan@slideshare.net(pelikan) Initial-Population Bias in the Univariate Estimation of Distribution Algorithm pelikan <img src="http://cdn.slidesharecdn.com/umda-bias-090713014644-phpapp02-thumbnail-2?1247467617" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> Initial-Population Bias in the Univariate Estimation of Distribution Algorithm
View more documents from pelikan.
]]> 468 0 http://cdn.slidesharecdn.com/umda-bias-090713014644-phpapp02-thumbnail-2?1247467617 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Using Previous Models to Bias Structural Learning in the Hierarchical BOA http://www.slideshare.net/pelikan/using-previous-models-to-bias-structural-learning-in-the-hierarchical-boa
Estimation of distribution algorithms (EDAs) are stochastic optimization techniques that explore the space of potential solutions by building and sampling explicit probabilistic models of promising candidate solutions. While the primary goal of applying EDAs is to discover the global optimum or at least its accurate approximation, besides this, any EDA provides us with a sequence of probabilistic models, which in most cases hold a great deal of information about the problem. Although using problem-specific knowledge has been shown to significantly improve performance of EDAs and other evolutionary algorithms, this readily available source of problem-specific information has been practically ignored by the EDA community. This paper takes the first step towards the use of probabilistic models obtained by EDAs to speed up the solution of similar problems in future. More specifically, we propose two approaches to biasing model building in the hierarchical Bayesian optimization algorithm (hBOA) based on knowledge automatically learned from previous hBOA runs on similar problems. We show that the proposed methods lead to substantial speedups and argue that the methods should work well in other applications that require solving a large number of problems with similar structure.]]>
Estimation of distribution algorithms (EDAs) are stochastic optimization techniques that explore the space of potential solutions by building and sampling explicit probabilistic models of promising candidate solutions. While the primary goal of applying EDAs is to discover the global optimum or at least its accurate approximation, besides this, any EDA provides us with a sequence of probabilistic models, which in most cases hold a great deal of information about the problem. Although using problem-specific knowledge has been shown to significantly improve performance of EDAs and other evolutionary algorithms, this readily available source of problem-specific information has been practically ignored by the EDA community. This paper takes the first step towards the use of probabilistic models obtained by EDAs to speed up the solution of similar problems in future. More specifically, we propose two approaches to biasing model building in the hierarchical Bayesian optimization algorithm (hBOA) based on knowledge automatically learned from previous hBOA runs on similar problems. We show that the proposed methods lead to substantial speedups and argue that the methods should work well in other applications that require solving a large number of problems with similar structure.]]> Tue, 22 Jul 2008 08:39:41 GMT http://www.slideshare.net/pelikan/using-previous-models-to-bias-structural-learning-in-the-hierarchical-boa pelikan@slideshare.net(pelikan) Using Previous Models to Bias Structural Learning in the Hierarchical BOA pelikan Estimation of distribution algorithms (EDAs) are stochastic optimization techniques that explore the space of potential solutions by building and sampling explicit probabilistic models of promising candidate solutions. While the primary goal of applying EDAs is to discover the global optimum or at least its accurate approximation, besides this, any EDA provides us with a sequence of probabilistic models, which in most cases hold a great deal of information about the problem. Although using problem-specific knowledge has been shown to significantly improve performance of EDAs and other evolutionary algorithms, this readily available source of problem-specific information has been practically ignored by the EDA community. This paper takes the first step towards the use of probabilistic models obtained by EDAs to speed up the solution of similar problems in future. More specifically, we propose two approaches to biasing model building in the hierarchical Bayesian optimization algorithm (hBOA) based on knowledge automatically learned from previous hBOA runs on similar problems. We show that the proposed methods lead to substantial speedups and argue that the methods should work well in other applications that require solving a large number of problems with similar structure. <img src="http://cdn.slidesharecdn.com/g08biashauschildfinal-1216723580077074-9-thumbnail-2?1216715983" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> Estimation of distribution algorithms (EDAs) are stochastic optimization techniques that explore the space of potential solutions by building and sampling explicit probabilistic models of promising candidate solutions. While the primary goal of applying EDAs is to discover the global optimum or at least its accurate approximation, besides this, any EDA provides us with a sequence of probabilistic models, which in most cases hold a great deal of information about the problem. Although using problem-specific knowledge has been shown to significantly improve performance of EDAs and other evolutionary algorithms, this readily available source of problem-specific information has been practically ignored by the EDA community. This paper takes the first step towards the use of probabilistic models obtained by EDAs to speed up the solution of similar problems in future. More specifically, we propose two approaches to biasing model building in the hierarchical Bayesian optimization algorithm (hBOA) based on knowledge automatically learned from previous hBOA runs on similar problems. We show that the proposed methods lead to substantial speedups and argue that the methods should work well in other applications that require solving a large number of problems with similar structure. Using Previous Models to Bias Structural Learning in the Hierarchical BOA
View more documents from pelikan.
]]> 931 0 http://cdn.slidesharecdn.com/g08biashauschildfinal-1216723580077074-9-thumbnail-2?1216715983 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Efficiency Enhancement of Estimation of Distribution Algorithms http://www.slideshare.net/pelikan/efficiency-enhancement-of-estimation-of-distribution-algorithms-518541
]]>
]]> Fri, 18 Jul 2008 12:03:26 GMT http://www.slideshare.net/pelikan/efficiency-enhancement-of-estimation-of-distribution-algorithms-518541 pelikan@slideshare.net(pelikan) Efficiency Enhancement of Estimation of Distribution Algorithms pelikan <img src="http://cdn.slidesharecdn.com/eeedapresentation-1216389990041960-9-thumbnail-2?1216382606" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> Efficiency Enhancement of Estimation of Distribution Algorithms
View more documents from pelikan.
]]> 1253 0 http://cdn.slidesharecdn.com/eeedapresentation-1216389990041960-9-thumbnail-2?1216382606 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Finding Ground States of Sherrington-Kirkpatrick Spin Glasses with Hierarchical BOA and Genetic Algorithms http://www.slideshare.net/pelikan/finding-ground-states-of-sherringtonkirkpatrick-spin-glasses-with-hierarchical-boa-and-genetic-algorithms
This study focuses on the problem of finding ground states of random instances of the Sherrington-Kirkpatrick (SK) spin-glass model with Gaussian couplings. While the ground states of SK spin-glass instances can be obtained with branch and bound, the computational complexity of branch and bound yields instances of not more than about 90 spins. We describe several approaches based on the hierarchical Bayesian optimization algorithm (hBOA) to reliably identifying ground states of SK instances intractable with branch and bound, and present a broad range of empirical results on such problem instances. We argue that the proposed methodology holds a big promise for reliably solving large SK spin-glass instances to optimality with practical time complexity. The proposed approaches to identifying global optima reliably can also be applied to other problems and they can be used with many other evolutionary algorithms. Performance of hBOA is compared to that of the genetic algorithm with two common crossover operators.]]>
This study focuses on the problem of finding ground states of random instances of the Sherrington-Kirkpatrick (SK) spin-glass model with Gaussian couplings. While the ground states of SK spin-glass instances can be obtained with branch and bound, the computational complexity of branch and bound yields instances of not more than about 90 spins. We describe several approaches based on the hierarchical Bayesian optimization algorithm (hBOA) to reliably identifying ground states of SK instances intractable with branch and bound, and present a broad range of empirical results on such problem instances. We argue that the proposed methodology holds a big promise for reliably solving large SK spin-glass instances to optimality with practical time complexity. The proposed approaches to identifying global optima reliably can also be applied to other problems and they can be used with many other evolutionary algorithms. Performance of hBOA is compared to that of the genetic algorithm with two common crossover operators.]]> Fri, 18 Jul 2008 12:02:42 GMT http://www.slideshare.net/pelikan/finding-ground-states-of-sherringtonkirkpatrick-spin-glasses-with-hierarchical-boa-and-genetic-algorithms pelikan@slideshare.net(pelikan) Finding Ground States of Sherrington-Kirkpatrick Spin Glasses with Hierarchical BOA and Genetic Algorithms pelikan This study focuses on the problem of finding ground states of random instances of the Sherrington-Kirkpatrick (SK) spin-glass model with Gaussian couplings. While the ground states of SK spin-glass instances can be obtained with branch and bound, the computational complexity of branch and bound yields instances of not more than about 90 spins. We describe several approaches based on the hierarchical Bayesian optimization algorithm (hBOA) to reliably identifying ground states of SK instances intractable with branch and bound, and present a broad range of empirical results on such problem instances. We argue that the proposed methodology holds a big promise for reliably solving large SK spin-glass instances to optimality with practical time complexity. The proposed approaches to identifying global optima reliably can also be applied to other problems and they can be used with many other evolutionary algorithms. Performance of hBOA is compared to that of the genetic algorithm with two common crossover operators. <img src="http://cdn.slidesharecdn.com/skspinglasspresentation-1216390089714696-9-thumbnail-2?1216602813" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> This study focuses on the problem of finding ground states of random instances of the Sherrington-Kirkpatrick (SK) spin-glass model with Gaussian couplings. While the ground states of SK spin-glass instances can be obtained with branch and bound, the computational complexity of branch and bound yields instances of not more than about 90 spins. We describe several approaches based on the hierarchical Bayesian optimization algorithm (hBOA) to reliably identifying ground states of SK instances intractable with branch and bound, and present a broad range of empirical results on such problem instances. We argue that the proposed methodology holds a big promise for reliably solving large SK spin-glass instances to optimality with practical time complexity. The proposed approaches to identifying global optima reliably can also be applied to other problems and they can be used with many other evolutionary algorithms. Performance of hBOA is compared to that of the genetic algorithm with two common crossover operators. Finding Ground States of Sherrington-Kirkpatrick Spin Glasses with Hierarchical BOA and Genetic Algorithms
View more documents from pelikan.
]]> 1039 0 http://cdn.slidesharecdn.com/skspinglasspresentation-1216390089714696-9-thumbnail-2?1216602813 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted iBOA: The Incremental Bayesian Optimization Algorithm http://www.slideshare.net/pelikan/iboa-the-incremental-bayesian-optimization-algorithm
This paper proposes the incremental Bayesian optimization algorithm (iBOA), which modifies standard BOA by removing the population of solutions and using incremental updates of the Bayesian network. iBOA is shown to be able to learn and exploit unrestricted Bayesian networks using incremental techniques for updating both the structure as well as the parameters of the probabilistic model. This represents an important step toward the design of competent incremental estimation of distribution algorithms that can solve difficult nearly decomposable problems scalably and reliably.]]>
This paper proposes the incremental Bayesian optimization algorithm (iBOA), which modifies standard BOA by removing the population of solutions and using incremental updates of the Bayesian network. iBOA is shown to be able to learn and exploit unrestricted Bayesian networks using incremental techniques for updating both the structure as well as the parameters of the probabilistic model. This represents an important step toward the design of competent incremental estimation of distribution algorithms that can solve difficult nearly decomposable problems scalably and reliably.]]> Fri, 18 Jul 2008 05:02:45 GMT http://www.slideshare.net/pelikan/iboa-the-incremental-bayesian-optimization-algorithm pelikan@slideshare.net(pelikan) iBOA: The Incremental Bayesian Optimization Algorithm pelikan This paper proposes the incremental Bayesian optimization algorithm (iBOA), which modifies standard BOA by removing the population of solutions and using incremental updates of the Bayesian network. iBOA is shown to be able to learn and exploit unrestricted Bayesian networks using incremental techniques for updating both the structure as well as the parameters of the probabilistic model. This represents an important step toward the design of competent incremental estimation of distribution algorithms that can solve difficult nearly decomposable problems scalably and reliably. <img src="http://cdn.slidesharecdn.com/iboapresentation-1216390049667260-9-thumbnail-2?1231920513" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> This paper proposes the incremental Bayesian optimization algorithm (iBOA), which modifies standard BOA by removing the population of solutions and using incremental updates of the Bayesian network. iBOA is shown to be able to learn and exploit unrestricted Bayesian networks using incremental techniques for updating both the structure as well as the parameters of the probabilistic model. This represents an important step toward the design of competent incremental estimation of distribution algorithms that can solve difficult nearly decomposable problems scalably and reliably. iBOA: The Incremental Bayesian Optimization Algorithm
View more documents from pelikan.
]]> 1582 0 http://cdn.slidesharecdn.com/iboapresentation-1216390049667260-9-thumbnail-2?1231920513 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Analysis of Estimation of Distribution Algorithms and Genetic Algorithms on NK Landscapes http://www.slideshare.net/pelikan/analysis-of-estimation-of-distribution-algorithms-and-genetic-algorithms-on-nk-landscapes
This study analyzes performance of several genetic and evolutionary algorithms on randomly generated NK fitness landscapes with various values of $n$ and $k$. A large number of NK problem instances are first generated for each $n$ and $k$, and the global optimum of each instance is obtained using the branch-and-bound algorithm. Next, the hierarchical Bayesian optimization algorithm (hBOA), the univariate marginal distribution algorithm (UMDA), and the simple genetic algorithm (GA) with uniform and two-point crossover operators are applied to all generated instances. Performance of all algorithms is then analyzed and compared, and the results are discussed.]]>
This study analyzes performance of several genetic and evolutionary algorithms on randomly generated NK fitness landscapes with various values of $n$ and $k$. A large number of NK problem instances are first generated for each $n$ and $k$, and the global optimum of each instance is obtained using the branch-and-bound algorithm. Next, the hierarchical Bayesian optimization algorithm (hBOA), the univariate marginal distribution algorithm (UMDA), and the simple genetic algorithm (GA) with uniform and two-point crossover operators are applied to all generated instances. Performance of all algorithms is then analyzed and compared, and the results are discussed.]]> Fri, 18 Jul 2008 05:02:38 GMT http://www.slideshare.net/pelikan/analysis-of-estimation-of-distribution-algorithms-and-genetic-algorithms-on-nk-landscapes pelikan@slideshare.net(pelikan) Analysis of Estimation of Distribution Algorithms and Genetic Algorithms on NK Landscapes pelikan This study analyzes performance of several genetic and evolutionary algorithms on randomly generated NK fitness landscapes with various values of $n$ and $k$. A large number of NK problem instances are first generated for each $n$ and $k$, and the global optimum of each instance is obtained using the branch-and-bound algorithm. Next, the hierarchical Bayesian optimization algorithm (hBOA), the univariate marginal distribution algorithm (UMDA), and the simple genetic algorithm (GA) with uniform and two-point crossover operators are applied to all generated instances. Performance of all algorithms is then analyzed and compared, and the results are discussed. <img src="http://cdn.slidesharecdn.com/nklandscapespresentation-1216390024451870-9-thumbnail-2?1231920512" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> This study analyzes performance of several genetic and evolutionary algorithms on randomly generated NK fitness landscapes with various values of $n$ and $k$. A large number of NK problem instances are first generated for each $n$ and $k$, and the global optimum of each instance is obtained using the branch-and-bound algorithm. Next, the hierarchical Bayesian optimization algorithm (hBOA), the univariate marginal distribution algorithm (UMDA), and the simple genetic algorithm (GA) with uniform and two-point crossover operators are applied to all generated instances. Performance of all algorithms is then analyzed and compared, and the results are discussed. Analysis of Estimation of Distribution Algorithms and Genetic Algorithms on NK Landscapes
View more documents from pelikan.
]]> 994 0 http://cdn.slidesharecdn.com/nklandscapespresentation-1216390024451870-9-thumbnail-2?1231920512 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Fitness inheritance in the Bayesian optimization algorithm http://www.slideshare.net/pelikan/fitness-inheritance-in-the-bayesian-optimization-algorithm
This paper describes how fitness inheritance can be used to estimate fitness for a proportion of newly sampled candidate solutions in the Bayesian optimization algorithm (BOA). The goal of estimating fitness for some candidate solutions is to reduce the number of fitness evaluations for problems where fitness evaluation is expensive. Bayesian networks used in BOA to model promising solutions and generate the new ones are extended to allow not only for modeling and sampling candidate solutions, but also for estimating their fitness. The results indicate that fitness inheritance is a promising concept in BOA, because population-sizing requirements for building appropriate models of promising solutions lead to good fitness estimates even if only a small proportion of candidate solutions is evaluated using the actual fitness function. This can lead to a reduction of the number of actual fitness evaluations by a factor of 30 or more.]]>
This paper describes how fitness inheritance can be used to estimate fitness for a proportion of newly sampled candidate solutions in the Bayesian optimization algorithm (BOA). The goal of estimating fitness for some candidate solutions is to reduce the number of fitness evaluations for problems where fitness evaluation is expensive. Bayesian networks used in BOA to model promising solutions and generate the new ones are extended to allow not only for modeling and sampling candidate solutions, but also for estimating their fitness. The results indicate that fitness inheritance is a promising concept in BOA, because population-sizing requirements for building appropriate models of promising solutions lead to good fitness estimates even if only a small proportion of candidate solutions is evaluated using the actual fitness function. This can lead to a reduction of the number of actual fitness evaluations by a factor of 30 or more.]]> Tue, 25 Sep 2007 11:57:08 GMT http://www.slideshare.net/pelikan/fitness-inheritance-in-the-bayesian-optimization-algorithm pelikan@slideshare.net(pelikan) Fitness inheritance in the Bayesian optimization algorithm pelikan This paper describes how fitness inheritance can be used to estimate fitness for a proportion of newly sampled candidate solutions in the Bayesian optimization algorithm (BOA). The goal of estimating fitness for some candidate solutions is to reduce the number of fitness evaluations for problems where fitness evaluation is expensive. Bayesian networks used in BOA to model promising solutions and generate the new ones are extended to allow not only for modeling and sampling candidate solutions, but also for estimating their fitness. The results indicate that fitness inheritance is a promising concept in BOA, because population-sizing requirements for building appropriate models of promising solutions lead to good fitness estimates even if only a small proportion of candidate solutions is evaluated using the actual fitness function. This can lead to a reduction of the number of actual fitness evaluations by a factor of 30 or more. <img src="http://cdn.slidesharecdn.com/fitness-inheritance-in-the-bayesian-optimization-algorithm943-thumbnail-2?1190721428" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> This paper describes how fitness inheritance can be used to estimate fitness for a proportion of newly sampled candidate solutions in the Bayesian optimization algorithm (BOA). The goal of estimating fitness for some candidate solutions is to reduce the number of fitness evaluations for problems where fitness evaluation is expensive. Bayesian networks used in BOA to model promising solutions and generate the new ones are extended to allow not only for modeling and sampling candidate solutions, but also for estimating their fitness. The results indicate that fitness inheritance is a promising concept in BOA, because population-sizing requirements for building appropriate models of promising solutions lead to good fitness estimates even if only a small proportion of candidate solutions is evaluated using the actual fitness function. This can lead to a reduction of the number of actual fitness evaluations by a factor of 30 or more. Fitness inheritance in the Bayesian optimization algorithm
View more documents from pelikan.
]]> 1107 0 http://cdn.slidesharecdn.com/fitness-inheritance-in-the-bayesian-optimization-algorithm943-thumbnail-2?1190721428 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Computational complexity and *simulation of rare events of Ising spin glasses http://www.slideshare.net/pelikan/computational-complexity-and-simulation-of-rare-events-of-ising-spin-glasses
We discuss the computational complexity of random 2D Ising spin glasses, which represent an interesting class of constraint satisfaction problems for black box optimization. Two extremal cases are considered: (1) the +/- J spin glass, and (2) the Gaussian spin glass. We also study a smooth transition between these two extremal cases. The computational complexity of all studied spin glass systems is found to be dominated by rare events of extremely hard spin glass samples. We show that complexity of all studied spin glass systems is closely related to Frechet extremal value distribution. In a hybrid algorithm that combines the hierarchical Bayesian optimization algorithm (hBOA) with a deterministic bit-flip hill climber, the number of steps performed by both the global searcher (hBOA) and the local searcher follow Frechet distributions. Nonetheless, unlike in methods based purely on local search, the parameters of these distributions confirm good scalability of hBOA with local search. We further argue that standard performance measures for optimization algorithms---such as the average number of evaluations until convergence---can be misleading. Finally, our results indicate that for highly multimodal constraint satisfaction problems, such as Ising spin glasses, recombination-based search can provide qualitatively better results than mutation-based search.]]>
We discuss the computational complexity of random 2D Ising spin glasses, which represent an interesting class of constraint satisfaction problems for black box optimization. Two extremal cases are considered: (1) the +/- J spin glass, and (2) the Gaussian spin glass. We also study a smooth transition between these two extremal cases. The computational complexity of all studied spin glass systems is found to be dominated by rare events of extremely hard spin glass samples. We show that complexity of all studied spin glass systems is closely related to Frechet extremal value distribution. In a hybrid algorithm that combines the hierarchical Bayesian optimization algorithm (hBOA) with a deterministic bit-flip hill climber, the number of steps performed by both the global searcher (hBOA) and the local searcher follow Frechet distributions. Nonetheless, unlike in methods based purely on local search, the parameters of these distributions confirm good scalability of hBOA with local search. We further argue that standard performance measures for optimization algorithms---such as the average number of evaluations until convergence---can be misleading. Finally, our results indicate that for highly multimodal constraint satisfaction problems, such as Ising spin glasses, recombination-based search can provide qualitatively better results than mutation-based search.]]> Tue, 25 Sep 2007 11:50:02 GMT http://www.slideshare.net/pelikan/computational-complexity-and-simulation-of-rare-events-of-ising-spin-glasses pelikan@slideshare.net(pelikan) Computational complexity and *simulation of rare events of Ising spin glasses pelikan We discuss the computational complexity of random 2D Ising spin glasses, which represent an interesting class of constraint satisfaction problems for black box optimization. Two extremal cases are considered: (1) the +/- J spin glass, and (2) the Gaussian spin glass. We also study a smooth transition between these two extremal cases. The computational complexity of all studied spin glass systems is found to be dominated by rare events of extremely hard spin glass samples. We show that complexity of all studied spin glass systems is closely related to Frechet extremal value distribution. In a hybrid algorithm that combines the hierarchical Bayesian optimization algorithm (hBOA) with a deterministic bit-flip hill climber, the number of steps performed by both the global searcher (hBOA) and the local searcher follow Frechet distributions. Nonetheless, unlike in methods based purely on local search, the parameters of these distributions confirm good scalability of hBOA with local search. We further argue that standard performance measures for optimization algorithms---such as the average number of evaluations until convergence---can be misleading. Finally, our results indicate that for highly multimodal constraint satisfaction problems, such as Ising spin glasses, recombination-based search can provide qualitatively better results than mutation-based search. <img src="http://cdn.slidesharecdn.com/computational-complexity-and-simulation-of-rare-events-of-ising-spin-glasses4173-thumbnail-2?1190721002" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> We discuss the computational complexity of random 2D Ising spin glasses, which represent an interesting class of constraint satisfaction problems for black box optimization. Two extremal cases are considered: (1) the +/- J spin glass, and (2) the Gaussian spin glass. We also study a smooth transition between these two extremal cases. The computational complexity of all studied spin glass systems is found to be dominated by rare events of extremely hard spin glass samples. We show that complexity of all studied spin glass systems is closely related to Frechet extremal value distribution. In a hybrid algorithm that combines the hierarchical Bayesian optimization algorithm (hBOA) with a deterministic bit-flip hill climber, the number of steps performed by both the global searcher (hBOA) and the local searcher follow Frechet distributions. Nonetheless, unlike in methods based purely on local search, the parameters of these distributions confirm good scalability of hBOA with local search. We further argue that standard performance measures for optimization algorithms---such as the average number of evaluations until convergence---can be misleading. Finally, our results indicate that for highly multimodal constraint satisfaction problems, such as Ising spin glasses, recombination-based search can provide qualitatively better results than mutation-based search. Computational complexity and simulation of rare events of Ising spin glasses
View more documents from pelikan.
]]> 1468 2 http://cdn.slidesharecdn.com/computational-complexity-and-simulation-of-rare-events-of-ising-spin-glasses4173-thumbnail-2?1190721002 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted The Bayesian Optimization Algorithm with Substructural Local Search http://www.slideshare.net/pelikan/the-bayesian-optimization-algorithm-with-substructural-local-search
This work studies the utility of using substructural neighborhoods for local search in the Bayesian optimization algorithm (BOA). The probabilistic model of BOA, which automatically identifies important problem substructures, is used to define the structure of the neighborhoods used in local search. Additionally, a surrogate fitness model is considered to evaluate the improvement of the local search steps. The results show that performing substructural local search in BOA significatively reduces the number of generations necessary to converge to optimal solutions and thus provides substantial speedups.]]>
This work studies the utility of using substructural neighborhoods for local search in the Bayesian optimization algorithm (BOA). The probabilistic model of BOA, which automatically identifies important problem substructures, is used to define the structure of the neighborhoods used in local search. Additionally, a surrogate fitness model is considered to evaluate the improvement of the local search steps. The results show that performing substructural local search in BOA significatively reduces the number of generations necessary to converge to optimal solutions and thus provides substantial speedups.]]> Wed, 19 Sep 2007 03:06:23 GMT http://www.slideshare.net/pelikan/the-bayesian-optimization-algorithm-with-substructural-local-search pelikan@slideshare.net(pelikan) The Bayesian Optimization Algorithm with Substructural Local Search pelikan This work studies the utility of using substructural neighborhoods for local search in the Bayesian optimization algorithm (BOA). The probabilistic model of BOA, which automatically identifies important problem substructures, is used to define the structure of the neighborhoods used in local search. Additionally, a surrogate fitness model is considered to evaluate the improvement of the local search steps. The results show that performing substructural local search in BOA significatively reduces the number of generations necessary to converge to optimal solutions and thus provides substantial speedups. <img src="http://cdn.slidesharecdn.com/the-bayesian-optimization-algorithm-with-substructural-local-search4950-thumbnail-2?1190171183" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> This work studies the utility of using substructural neighborhoods for local search in the Bayesian optimization algorithm (BOA). The probabilistic model of BOA, which automatically identifies important problem substructures, is used to define the structure of the neighborhoods used in local search. Additionally, a surrogate fitness model is considered to evaluate the improvement of the local search steps. The results show that performing substructural local search in BOA significatively reduces the number of generations necessary to converge to optimal solutions and thus provides substantial speedups. The Bayesian Optimization Algorithm with Substructural Local Search
View more documents from pelikan.
]]> 1746 0 http://cdn.slidesharecdn.com/the-bayesian-optimization-algorithm-with-substructural-local-search4950-thumbnail-2?1190171183 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Analyzing Probabilistic Models in Hierarchical BOA on Traps and Spin Glasses http://www.slideshare.net/pelikan/analyzing-probabilistic-models-in-hierarchical-boa-on-traps-and-spin-glasses-83852
The hierarchical Bayesian optimization algorithm (hBOA) can solve nearly decomposable and hierarchical problems of bounded difficulty in a robust and scalable manner by building and sampling probabilistic models of promising solutions. This paper analyzes probabilistic models in hBOA on two common test problems: concatenated traps and 2D Ising spin glasses with periodic boundary conditions. We argue that although Bayesian networks with local structures can encode complex probability distributions, analyzing these models in hBOA is relatively straightforward and the results of such analyses may provide practitioners with useful information about their problems. The results show that the probabilistic models in hBOA closely correspond to the structure of the underlying optimization problem, the models do not change significantly in subsequent iterations of BOA, and creating adequate probabilistic models by hand is not straightforward even with complete knowledge of the optimization problem.]]>
The hierarchical Bayesian optimization algorithm (hBOA) can solve nearly decomposable and hierarchical problems of bounded difficulty in a robust and scalable manner by building and sampling probabilistic models of promising solutions. This paper analyzes probabilistic models in hBOA on two common test problems: concatenated traps and 2D Ising spin glasses with periodic boundary conditions. We argue that although Bayesian networks with local structures can encode complex probability distributions, analyzing these models in hBOA is relatively straightforward and the results of such analyses may provide practitioners with useful information about their problems. The results show that the probabilistic models in hBOA closely correspond to the structure of the underlying optimization problem, the models do not change significantly in subsequent iterations of BOA, and creating adequate probabilistic models by hand is not straightforward even with complete knowledge of the optimization problem.]]> Sat, 28 Jul 2007 11:23:25 GMT http://www.slideshare.net/pelikan/analyzing-probabilistic-models-in-hierarchical-boa-on-traps-and-spin-glasses-83852 pelikan@slideshare.net(pelikan) Analyzing Probabilistic Models in Hierarchical BOA on Traps and Spin Glasses pelikan The hierarchical Bayesian optimization algorithm (hBOA) can solve nearly decomposable and hierarchical problems of bounded difficulty in a robust and scalable manner by building and sampling probabilistic models of promising solutions. This paper analyzes probabilistic models in hBOA on two common test problems: concatenated traps and 2D Ising spin glasses with periodic boundary conditions. We argue that although Bayesian networks with local structures can encode complex probability distributions, analyzing these models in hBOA is relatively straightforward and the results of such analyses may provide practitioners with useful information about their problems. The results show that the probabilistic models in hBOA closely correspond to the structure of the underlying optimization problem, the models do not change significantly in subsequent iterations of BOA, and creating adequate probabilistic models by hand is not straightforward even with complete knowledge of the optimization problem. <img src="http://cdn.slidesharecdn.com/analyzing-probabilistic-models-in-hierarchical-boa-on-traps-and-spin-glasses574-thumbnail-2?1185621805" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> The hierarchical Bayesian optimization algorithm (hBOA) can solve nearly decomposable and hierarchical problems of bounded difficulty in a robust and scalable manner by building and sampling probabilistic models of promising solutions. This paper analyzes probabilistic models in hBOA on two common test problems: concatenated traps and 2D Ising spin glasses with periodic boundary conditions. We argue that although Bayesian networks with local structures can encode complex probability distributions, analyzing these models in hBOA is relatively straightforward and the results of such analyses may provide practitioners with useful information about their problems. The results show that the probabilistic models in hBOA closely correspond to the structure of the underlying optimization problem, the models do not change significantly in subsequent iterations of BOA, and creating adequate probabilistic models by hand is not straightforward even with complete knowledge of the optimization problem. Analyzing Probabilistic Models in Hierarchical BOA on Traps and Spin Glasses
View more documents from pelikan.
]]> 1324 0 http://cdn.slidesharecdn.com/analyzing-probabilistic-models-in-hierarchical-boa-on-traps-and-spin-glasses574-thumbnail-2?1185621805 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Hybrid Evolutionary Algorithms on Minimum Vertex Cover for Random Graphs http://www.slideshare.net/pelikan/hybrid-evolutionary-algorithms-on-minimum-vertex-cover-for-random-graphs
This work analyzes the hierarchical Bayesian optimization algorithm (hBOA) on minimum vertex cover for standard classes of random graphs and transformed SAT instances. The performance of hBOA is compared with that of the branch-and-bound problem solver (BB), the simple genetic algorithm (GA) and the parallel simulated annealing (PSA). The results indicate that BB is significantly outperformed by all the other tested methods, which is expected as BB is a complete search algorithm and minimum vertex cover is an NP-complete problem. The best performance is achieved by hBOA; nonetheless, the performance differences between hBOA and other evolutionary algorithms are relatively small, indicating that mutation-based search and recombination-based search lead to similar performance on the tested classes of minimum vertex cover problems.]]>
This work analyzes the hierarchical Bayesian optimization algorithm (hBOA) on minimum vertex cover for standard classes of random graphs and transformed SAT instances. The performance of hBOA is compared with that of the branch-and-bound problem solver (BB), the simple genetic algorithm (GA) and the parallel simulated annealing (PSA). The results indicate that BB is significantly outperformed by all the other tested methods, which is expected as BB is a complete search algorithm and minimum vertex cover is an NP-complete problem. The best performance is achieved by hBOA; nonetheless, the performance differences between hBOA and other evolutionary algorithms are relatively small, indicating that mutation-based search and recombination-based search lead to similar performance on the tested classes of minimum vertex cover problems.]]> Thu, 26 Jul 2007 05:45:31 GMT http://www.slideshare.net/pelikan/hybrid-evolutionary-algorithms-on-minimum-vertex-cover-for-random-graphs pelikan@slideshare.net(pelikan) Hybrid Evolutionary Algorithms on Minimum Vertex Cover for Random Graphs pelikan This work analyzes the hierarchical Bayesian optimization algorithm (hBOA) on minimum vertex cover for standard classes of random graphs and transformed SAT instances. The performance of hBOA is compared with that of the branch-and-bound problem solver (BB), the simple genetic algorithm (GA) and the parallel simulated annealing (PSA). The results indicate that BB is significantly outperformed by all the other tested methods, which is expected as BB is a complete search algorithm and minimum vertex cover is an NP-complete problem. The best performance is achieved by hBOA; nonetheless, the performance differences between hBOA and other evolutionary algorithms are relatively small, indicating that mutation-based search and recombination-based search lead to similar performance on the tested classes of minimum vertex cover problems. <img src="http://cdn.slidesharecdn.com/hybrid-evolutionary-algorithms-on-minimum-vertex-cover-for-random-graphs3137-thumbnail-2?1185428731" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> This work analyzes the hierarchical Bayesian optimization algorithm (hBOA) on minimum vertex cover for standard classes of random graphs and transformed SAT instances. The performance of hBOA is compared with that of the branch-and-bound problem solver (BB), the simple genetic algorithm (GA) and the parallel simulated annealing (PSA). The results indicate that BB is significantly outperformed by all the other tested methods, which is expected as BB is a complete search algorithm and minimum vertex cover is an NP-complete problem. The best performance is achieved by hBOA; nonetheless, the performance differences between hBOA and other evolutionary algorithms are relatively small, indicating that mutation-based search and recombination-based search lead to similar performance on the tested classes of minimum vertex cover problems. Hybrid Evolutionary Algorithms on Minimum Vertex Cover for Random Graphs
View more documents from pelikan.
]]> 1715 0 http://cdn.slidesharecdn.com/hybrid-evolutionary-algorithms-on-minimum-vertex-cover-for-random-graphs3137-thumbnail-2?1185428731 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Order Or Not: Does Parallelization of Model Building in hBOA Affect Its Scalability? http://www.slideshare.net/pelikan/order-or-not-does-parallelization-of-model-building-in-hboa-affect-its-scalability
It has been shown that model building in the hierarchical Bayesian optimization algorithm (hBOA) can be efficiently parallelized by randomly generating an ancestral ordering of the nodes of the network prior to learning the network structure and allowing only dependencies consistent with the generated ordering. However, it has not been thoroughly shown that this approach to restricting probabilistic models does not affect scalability of hBOA on important classes of problems. This presentation demonstrates that although the use of a random ancestral ordering restricts the structure of considered models to allow efficient parallelization of model building, its effects on hBOA performance and scalability are negligible.]]>
It has been shown that model building in the hierarchical Bayesian optimization algorithm (hBOA) can be efficiently parallelized by randomly generating an ancestral ordering of the nodes of the network prior to learning the network structure and allowing only dependencies consistent with the generated ordering. However, it has not been thoroughly shown that this approach to restricting probabilistic models does not affect scalability of hBOA on important classes of problems. This presentation demonstrates that although the use of a random ancestral ordering restricts the structure of considered models to allow efficient parallelization of model building, its effects on hBOA performance and scalability are negligible.]]> Thu, 26 Jul 2007 05:23:22 GMT http://www.slideshare.net/pelikan/order-or-not-does-parallelization-of-model-building-in-hboa-affect-its-scalability pelikan@slideshare.net(pelikan) Order Or Not: Does Parallelization of Model Building in hBOA Affect Its Scalability? pelikan It has been shown that model building in the hierarchical Bayesian optimization algorithm (hBOA) can be efficiently parallelized by randomly generating an ancestral ordering of the nodes of the network prior to learning the network structure and allowing only dependencies consistent with the generated ordering. However, it has not been thoroughly shown that this approach to restricting probabilistic models does not affect scalability of hBOA on important classes of problems. This presentation demonstrates that although the use of a random ancestral ordering restricts the structure of considered models to allow efficient parallelization of model building, its effects on hBOA performance and scalability are negligible. <img src="http://cdn.slidesharecdn.com/order-or-not-does-parallelization-of-model-building-in-hboa-affect-its-scalability1657-thumbnail-2?1185427402" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> It has been shown that model building in the hierarchical Bayesian optimization algorithm (hBOA) can be efficiently parallelized by randomly generating an ancestral ordering of the nodes of the network prior to learning the network structure and allowing only dependencies consistent with the generated ordering. However, it has not been thoroughly shown that this approach to restricting probabilistic models does not affect scalability of hBOA on important classes of problems. This presentation demonstrates that although the use of a random ancestral ordering restricts the structure of considered models to allow efficient parallelization of model building, its effects on hBOA performance and scalability are negligible. Order Or Not: Does Parallelization of Model Building in hBOA Affect Its Scalability?
View more documents from pelikan.
]]> 841 0 http://cdn.slidesharecdn.com/order-or-not-does-parallelization-of-model-building-in-hboa-affect-its-scalability1657-thumbnail-2?1185427402 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted Estimation of Distribution Algorithms Tutorial http://www.slideshare.net/pelikan/estimation-of-distribution-algorithms-tutorial
Probabilistic model-building algorithms (PMBGAs), also called estimation of distribution algorithms (EDAs) and iterated density estimation algorithms (IDEAs), replace traditional variation of genetic and evolutionary algorithms by (1) building a probabilistic model of promising solutions and (2) sampling the built model to generate new candidate solutions. PMBGAs are also known as estimation of distribution algorithms (EDAs) and iterated density-estimation algorithms (IDEAs). Replacing traditional crossover and mutation operators by building and sampling a probabilistic model of promising solutions enables the use of machine learning techniques for automatic discovery of problem regularities and exploitation of these regularities for effective exploration of the search space. Using machine learning in optimization enables the design of optimization techniques that can automatically adapt to the given problem. There are many successful applications of PMBGAs, for example, Ising spin glasses in 2D and 3D, graph partitioning, MAXSAT, feature subset selection, forest management, groundwater remediation design, telecommunication network design, antenna design, and scheduling. This tutorial provides a gentle introduction to PMBGAs with an overview of major research directions in this area. Strengths and weaknesses of different PMBGAs will be discussed and suggestions will be provided to help practitioners to choose the best PMBGA for their problem. The video of this tutorial presented at GECCO-2008 can be found at http://medal.cs.umsl.edu/blog/?p=293]]>
Probabilistic model-building algorithms (PMBGAs), also called estimation of distribution algorithms (EDAs) and iterated density estimation algorithms (IDEAs), replace traditional variation of genetic and evolutionary algorithms by (1) building a probabilistic model of promising solutions and (2) sampling the built model to generate new candidate solutions. PMBGAs are also known as estimation of distribution algorithms (EDAs) and iterated density-estimation algorithms (IDEAs). Replacing traditional crossover and mutation operators by building and sampling a probabilistic model of promising solutions enables the use of machine learning techniques for automatic discovery of problem regularities and exploitation of these regularities for effective exploration of the search space. Using machine learning in optimization enables the design of optimization techniques that can automatically adapt to the given problem. There are many successful applications of PMBGAs, for example, Ising spin glasses in 2D and 3D, graph partitioning, MAXSAT, feature subset selection, forest management, groundwater remediation design, telecommunication network design, antenna design, and scheduling. This tutorial provides a gentle introduction to PMBGAs with an overview of major research directions in this area. Strengths and weaknesses of different PMBGAs will be discussed and suggestions will be provided to help practitioners to choose the best PMBGA for their problem. The video of this tutorial presented at GECCO-2008 can be found at http://medal.cs.umsl.edu/blog/?p=293]]> Thu, 26 Jul 2007 05:20:36 GMT http://www.slideshare.net/pelikan/estimation-of-distribution-algorithms-tutorial pelikan@slideshare.net(pelikan) Estimation of Distribution Algorithms Tutorial pelikan Probabilistic model-building algorithms (PMBGAs), also called estimation of distribution algorithms (EDAs) and iterated density estimation algorithms (IDEAs), replace traditional variation of genetic and evolutionary algorithms by (1) building a probabilistic model of promising solutions and (2) sampling the built model to generate new candidate solutions. PMBGAs are also known as estimation of distribution algorithms (EDAs) and iterated density-estimation algorithms (IDEAs). Replacing traditional crossover and mutation operators by building and sampling a probabilistic model of promising solutions enables the use of machine learning techniques for automatic discovery of problem regularities and exploitation of these regularities for effective exploration of the search space. Using machine learning in optimization enables the design of optimization techniques that can automatically adapt to the given problem. There are many successful applications of PMBGAs, for example, Ising spin glasses in 2D and 3D, graph partitioning, MAXSAT, feature subset selection, forest management, groundwater remediation design, telecommunication network design, antenna design, and scheduling. This tutorial provides a gentle introduction to PMBGAs with an overview of major research directions in this area. Strengths and weaknesses of different PMBGAs will be discussed and suggestions will be provided to help practitioners to choose the best PMBGA for their problem. The video of this tutorial presented at GECCO-2008 can be found at http://medal.cs.umsl.edu/blog/?p=293 <img src="http://cdn.slidesharecdn.com/estimation-of-distribution-algorithms-tutorial1694-thumbnail-2?1239642838" alt ="" style="border:1px solid #C3E6D8;float:right;" /><br> Probabilistic model-building algorithms (PMBGAs), also called estimation of distribution algorithms (EDAs) and iterated density estimation algorithms (IDEAs), replace traditional variation of genetic and evolutionary algorithms by (1) building a probabilistic model of promising solutions and (2) sampling the built model to generate new candidate solutions. PMBGAs are also known as estimation of distribution algorithms (EDAs) and iterated density-estimation algorithms (IDEAs). Replacing traditional crossover and mutation operators by building and sampling a probabilistic model of promising solutions enables the use of machine learning techniques for automatic discovery of problem regularities and exploitation of these regularities for effective exploration of the search space. Using machine learning in optimization enables the design of optimization techniques that can automatically adapt to the given problem. There are many successful applications of PMBGAs, for example, Ising spin glasses in 2D and 3D, graph partitioning, MAXSAT, feature subset selection, forest management, groundwater remediation design, telecommunication network design, antenna design, and scheduling. This tutorial provides a gentle introduction to PMBGAs with an overview of major research directions in this area. Strengths and weaknesses of different PMBGAs will be discussed and suggestions will be provided to help practitioners to choose the best PMBGA for their problem. The video of this tutorial presented at GECCO-2008 can be found at http://medal.cs.umsl.edu/blog/?p=293 Estimation of Distribution Algorithms Tutorial
View more documents from pelikan.
]]> 5303 1 http://cdn.slidesharecdn.com/estimation-of-distribution-algorithms-tutorial1694-thumbnail-2?1239642838 presentation http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted
SlideShare | Get your SlideShare Playlist
]]>
SlideShare | Get your SlideShare Playlist
]]>
SlideShare | Get your Presentation Pack
]]>
SlideShare | Get your Presentation Pack
]]>
Get your own Widget
]]> http://cdn.slidesharecdn.com/profile-photo-pelikan?1254979349 I&rsquo;m Martin Pelikan, computer science professor and director of the Missouri Estimation of Distribution Algorithms Laboratory (MEDAL) at the University of Missouri in St. Louis. My research focuses on evolutionary computation and machine learning. medal.cs.umsl.edu/ http://cdn.slidesharecdn.com/dhc-hboa-090721222318-phpapp02-thumbnail?1248233013 pelikan/effects-of-a-deterministic-hill-climber-on-hboa Effects of a Determini... http://cdn.slidesharecdn.com/intelligent-bias-090721222200-phpapp01-thumbnail?1248233230 pelikan/ntelligent-bias-of-network-structures-in-the-hierarchical-boa Intelligent Bias of Ne... http://cdn.slidesharecdn.com/sk-ring-090713101332-phpapp01-thumbnail?1247498036 pelikan/analysis-of-evolutionary-algorithms-on-the-onedimensional-spin-glass-with-powerlaw-interactions Analysis of Evolutiona...