Ender Özcan - Profile on Academia.edu (original) (raw)

Papers by Ender Özcan

Johannes Ostler, 447 Barry O'Sullivan, 569

Hyper-heuristics are a class of high-level search techniques which operate on a search space of h... more Hyper-heuristics are a class of high-level search techniques which operate on a search space of heuristics rather than directly on a search space of solutions. Early hyperheuristics focussed on selecting and applying a low-level heuristic at each stage of a search. Recent trends in hyper-heuristic research have led to a number of approaches being developed to automatically generate new heuristics from a set of heuristic components. This work investigates the suitability of using genetic programming as a hyper-heuristic methodology to generate constructive heuristics to solve the multidimensional 0-1 knapsack problem. A population of heuristics to rank knapsack items are trained on a subset of test problems and then applied to unseen instances. The results over a set of standard benchmarks show that genetic programming can be used to generate constructive heuristics which yield human-competitive results.

Advances in Intelligent Systems and Computing, 2015

Hyper-heuristics are a class of high-level search methods used to solve computationally difficult... more Hyper-heuristics are a class of high-level search methods used to solve computationally difficult problems, which operate on a search space of low-level heuristics rather than solutions directly. Previous work has shown that selection hyper-heuristics are able to solve many combinatorial optimisation problems, including the multidimensional 0-1 knapsack problem (MKP). The traditional framework for iterative selection hyper-heuristics relies on two key components, a heuristic selection method and a move acceptance criterion. Existing work has shown that a hyper-heuristic using Modified Choice Function heuristic selection can be effective at solving problems in multiple problem domains. Late Acceptance Strategy is a hill climbing metaheuristic strategy often used as a move acceptance criteria in selection hyper-heuristics. This work compares a Modified Choice Function -Late Acceptance Strategy hyper-heuristic to an existing selection hyper-heuristic method from the literature which has previously performed well on standard MKP benchmarks.

Shape Recognition Using Genetic Algorithm

International Conference on Evolutionary Computation, 1996

Shape recognition is a challenging task when shapesoverlap, forming noisy, occluded, partial shap... more Shape recognition is a challenging task when shapesoverlap, forming noisy, occluded, partial shapes.This paper uses a genetic algorithm for matchinginput shapes with model shapes described in termsof features such as line segments and angles (extractedusing traditional algorithms). The qualityof matching is gauged using a measure derived fromattributed shape grammars [12, 13]. Preliminary results,using shapes with about 30 features each, areextremely

Proceedings of the 2014 conference companion on Genetic and evolutionary computation companion - GECCO Comp '14, 2014

The mutation operator is the only genetic operator in Evolutionary Programming (EP). In the past ... more The mutation operator is the only genetic operator in Evolutionary Programming (EP). In the past researchers have nominated Gaussian, Cauchy, and Lévy distributions as mutation operators. According to the No Free Lunch theorem , no single mutation operator is able to outperform all others over the set of all possible functions. Potentially there is a lot of useful information generated when EP is ongoing. In this paper, we collect such information and propose a step size based self-adaptive mutation operator for Evolutionary Programming (SSEP). In SSEP, the mutation operator might be changed during the evolutionary process, based on the step size, from generation to generation. Principles for selecting an appropriate mutation operator for EP is proposed, with SSEP grounded on the principles. SSEP is shown to outperform static mutation operators in Evolutionary Programming on most of the functions tested. We also compare the experimental results of SSEP with other recent Evolutionary Programming methods, which uses multiple mutation operators.

There is a variety of approaches developed by researchers to solve different instances of timetab... more There is a variety of approaches developed by researchers to solve different instances of timetabling problems. During these studies different data formats are used to represent a timetabling problem instance and its solution, causing difficulties in the evaluation and comparison of approaches and sharing data. In this paper, a model for timetabling problems and a new XML data format for

Computer and Information Sciences III, 2012

Previous work on the Daily Car-Pooling problem includes an algorithm that consists of greedy assi... more Previous work on the Daily Car-Pooling problem includes an algorithm that consists of greedy assignment alternating with random perturbation. In this study, we examine the effect of varying the move acceptance policy, specifically Late-acceptance criteria with and without reheating. Late acceptance-based move acceptance criteria were chosen because there is strong empirical evidence in the literature indicating their superiority. Late-acceptance compares the objective values of the current solution with one which was obtained at a fixed number of steps prior to the current step during the search process in order to make an acceptance decision. We observe that the Late-acceptance criteria also achieve superior results in over 75% of cases for the Daily Car-Pooling problem, the majority of these results being statistically significant.

2012 12th UK Workshop on Computational Intelligence (UKCI), 2012

An iterative selection hyper-heuristic method controls and mixes a set of low-level heuristics wh... more An iterative selection hyper-heuristic method controls and mixes a set of low-level heuristics while solving a given problem. A low-level heuristic is selected and employed for improving a (set of) solution(s) at each step. This study investigates the influence of different heuristic selection methods within a population based incremental learning algorithm and hyper-heuristic based hybrid multiphase framework for solving dynamic environment problems. Even though the hybrid method delivers a good overall performance, it is superior in cyclic environments. The empirical results show that a heuristic selection method that relies on a fixed permutation of the underlying low-level heuristics, combined with a strategy that guarantees diversity when the environment changes is more successful than the learning approaches across different cyclic dynamic environments produced by a well known benchmark generator.

Proceedings of the 11th annual conference companion on Genetic and evolutionary computation conference - GECCO '09, 2009

If an optimisation algorithm performs a search in an environment that changes over time, it shoul... more If an optimisation algorithm performs a search in an environment that changes over time, it should be able to follow these changes and adapt itself for handling them in order to achieve good results. Different types of dynamics in a changing environment require the use of different approaches. Hyper-heuristics represent a class of methodologies that are high level heuristics performing search over a set of low level heuristics. Due to the generality of hyperheuristic frameworks, they are expected to be adaptive. Hence, a hyper-heuristic can be used in a dynamic environment to determine the approach to apply, adapting itself accordingly at each change. This study presents an initial investigation of hyper-heuristics in dynamic environments. A greedy hyper-heuristic is tested over a set of benchmark functions.

Lecture Notes in Computer Science, 2013

The vehicle routing problem (VRP) is a family of problems whereby a fleet of vehicles must servic... more The vehicle routing problem (VRP) is a family of problems whereby a fleet of vehicles must service the commodity demands of a set of geographically scattered customers from one or more depots, subject to a number of constraints. Early hyper-heuristic research focussed on selecting and applying a low-level heuristic at a given stage of an optimisation process. Recent trends have led to a number of approaches being developed to automatically generate heuristics for a number of combinatorial optimisation problems. Previous work on the VRP has shown that the application of hyper-heuristic approaches can yield successful results. In this paper we investigate the potential of grammatical evolution as a method to evolve the components of a variable neighbourhood search (VNS) framework. In particular two components are generated; constructive heuristics to create initial solutions and neighbourhood move operators to change the state of a given solution. The proposed method is tested on standard benchmark instances of two common VRP variants.

Lecture Notes in Computer Science, 2008

Genetic art is a recent art form generated by computers based on the genetic algorithms (GAs). In... more Genetic art is a recent art form generated by computers based on the genetic algorithms (GAs). In this paper, the components of a GA embedded into a genetic art tool named AMUSE are introduced. AMUSE is used to generate improvised melodies over a musical piece given a harmonic context. Population of melodies is evolved towards a better musical form based on a fitness function that evaluates ten different melodic and rhythmic features. Performance analysis of the GA based on a public evaluation shows that the objectives used by the fitness function are assembled properly and it is a successful artificial intelligence application.

Lecture Notes in Computer Science, 2007

Optimization in sports is a field of increasing interest. Combinatorial optimization techniques h... more Optimization in sports is a field of increasing interest. Combinatorial optimization techniques have been applied e.g. to game scheduling and playoff elimination. A common problem usually found in sports management is the assignment of referees to games already scheduled. There are a number of rules and objectives that should be taken into account when referees are assigned to games. We address a simplified version of a referee assignment problem common to many amateur leagues of sports such as soccer, baseball, and basketball. The problem is formulated by integer programming and its decision version is proved to be NP-complete. To tackle real-life large instances of the referee assignment problem, we propose a three-phase heuristic approach based on a constructive procedure, a repair heuristic to make solutions feasible, and a local search heuristic to improve feasible solutions. Numerical results on realistic instances are presented and discussed.

Information Sciences, 2015

Hyper-heuristics have emerged as automated high level search methodologies that manage a set of l... more Hyper-heuristics have emerged as automated high level search methodologies that manage a set of low level heuristics for solving computationally hard problems. A generic selection hyper-heuristic combines heuristic selection and move acceptance methods under an iterative single point-based search framework. At each step, the solution in hand is modified after applying a selected heuristic and a decision is made whether the new solution is accepted or not. In this study, we represent the trail of a hyper-heuristic as a third order tensor. Factorization of such a tensor reveals the latent relationships between the low level heuristics and the hyper-heuristic itself. The proposed learning approach partitions the set of low level heuristics into two equal subsets where heuristics in each subset are associated with a separate move acceptance method. Then a multi-stage hyper-heuristic is formed and while solving a given problem instance, heuristics are allowed to operate only in conjunction with the associated acceptance method at each stage. To the best of our knowledge, this is the first time tensor analysis of the space of heuristics is used as a data science approach to improve the performance of a hyper-heuristic in the prescribed manner. The empirical results across six different problem domains from a benchmark indeed indicate the success of the proposed approach.

Soft morphological filter optimization using a genetic algorithm for noise elimination

2014 14th UK Workshop on Computational Intelligence (UKCI), 2014

2014 IEEE Symposium on Evolving and Autonomous Learning Systems (EALS), 2014

Apprenticeship learning occurs via observations while an expert is in action. A hyper-heuristic i... more Apprenticeship learning occurs via observations while an expert is in action. A hyper-heuristic is a search method or a learning mechanism that controls a set of low level heuristics or combines different heuristic components to generate heuristics for solving a given computationally hard problem. In this study, we investigate into a novel apprenticeshiplearning-based approach which is used to automatically generate a hyper-heuristic for vehicle routing. This approach itself can be considered as a hyper-heuristic which operates in a train and test fashion. A state-of-the-art hyper-heuristic is chosen as an expert which is the winner of a previous hyper-heuristic competition. Trained on small vehicle routing instances, the learning approach yields various classifiers, each capturing different actions that the expert hyper-heuristic performs during the search process. Those classifiers are then used to produce a hyper-heuristic which is potentially capable of generalizing the actions of the expert hyperheuristic while solving the unseen instances. The experimental results on vehicle routing using the Hyper-heuristic Flexible (HyFlex) framework shows that the apprenticeship-learningbased hyper-heuristic delivers an outstanding performance when compared to the expert and some other previously proposed hyper-heuristics.

Multidisciplinary Scheduling: Theory and Applications, 2005

There is a variety of approaches developed by researchers to solve different instances of timetab... more There is a variety of approaches developed by researchers to solve different instances of timetabling problems. During these studies different data formats are used to represent a timetabling problem instance and its solution, causing difficulties in the evaluation and comparison of approaches and sharing data. In this paper, a model for timetabling problems and a new XML data format for them based on MathML is proposed.

A grouping hyper-heuristic framework: Application on graph colouring

Expert Systems with Applications, 2015

ABSTRACT Grouping problems are hard to solve combinatorial optimisation problems which require pa... more ABSTRACT Grouping problems are hard to solve combinatorial optimisation problems which require partitioning of objects into a minimum number of subsets while a given objective is simultaneously optimised. Selection hyper-heuristics are high level general purpose search methodologies that operate on a space formed by a set of low level heuristics rather than solutions. Most of the recently proposed selection hyper-heuristics are iterative and make use of two key methods which are employed successively; heuristic selection and move acceptance.

Detecting change and dealing with uncertainty in imperfect evolutionary environments

Information Sciences, 2015

ABSTRACT Imperfection of information is a part of our daily life; however, it is usually ignored ... more ABSTRACT Imperfection of information is a part of our daily life; however, it is usually ignored in learning based on evolutionary approaches. In this paper we develop an Imperfect Evolutionary System that provides an uncertain and chaotic imperfect environment that presents new challenges to its habitants. We then propose an intelligent methodology which is capable of learning in such environments. Detecting changes and adapting to the new environment is crucial to exploring the search space and exploiting any new opportunities that may arise. To deal with these uncertain and challenging environments, we propose a novel change detection strategy based on a Particle Swarm Optimization system which is hybridized with an Artificial Neural Network. This approach maintains a balance between exploitation and exploration during the search process. A comparison of approaches using different Particle Swarm Optimization algorithms show that the ability of our learning approach to detect changes and adapt as per the new demands of the environment is high.

International Journal of Parallel Programming, 2006

Determining the optimum data distribution, degree of parallelism and the communication structure ... more Determining the optimum data distribution, degree of parallelism and the communication structure on Distributed Memory machines for a given algorithm is not a straightforward task. Assuming that a parallel algorithm consists of consecutive stages, a Genetic Algorithm is proposed to find the best number of processors and the best data distribution method to be used for each stage of the parallel algorithm. Steady state genetic algorithm is compared with transgenerational genetic algorithm using different crossover operators. Performance is evaluated in terms of the total execution time of the program including communication and computation times. A computation intensive, a communication intensive and a mixed implementation are utilized in the experiments. The performance of GA provides satisfactory results for these illustrative examples.

A multi-level search framework for asynchronous cooperation of multiple hyper-heuristics

Proceedings of the 11th annual conference companion on Genetic and evolutionary computation conference - GECCO '09, 2009

ABSTRACT

Johannes Ostler, 447 Barry O'Sullivan, 569

Hyper-heuristics are a class of high-level search techniques which operate on a search space of h... more Hyper-heuristics are a class of high-level search techniques which operate on a search space of heuristics rather than directly on a search space of solutions. Early hyperheuristics focussed on selecting and applying a low-level heuristic at each stage of a search. Recent trends in hyper-heuristic research have led to a number of approaches being developed to automatically generate new heuristics from a set of heuristic components. This work investigates the suitability of using genetic programming as a hyper-heuristic methodology to generate constructive heuristics to solve the multidimensional 0-1 knapsack problem. A population of heuristics to rank knapsack items are trained on a subset of test problems and then applied to unseen instances. The results over a set of standard benchmarks show that genetic programming can be used to generate constructive heuristics which yield human-competitive results.

Advances in Intelligent Systems and Computing, 2015

Hyper-heuristics are a class of high-level search methods used to solve computationally difficult... more Hyper-heuristics are a class of high-level search methods used to solve computationally difficult problems, which operate on a search space of low-level heuristics rather than solutions directly. Previous work has shown that selection hyper-heuristics are able to solve many combinatorial optimisation problems, including the multidimensional 0-1 knapsack problem (MKP). The traditional framework for iterative selection hyper-heuristics relies on two key components, a heuristic selection method and a move acceptance criterion. Existing work has shown that a hyper-heuristic using Modified Choice Function heuristic selection can be effective at solving problems in multiple problem domains. Late Acceptance Strategy is a hill climbing metaheuristic strategy often used as a move acceptance criteria in selection hyper-heuristics. This work compares a Modified Choice Function -Late Acceptance Strategy hyper-heuristic to an existing selection hyper-heuristic method from the literature which has previously performed well on standard MKP benchmarks.

Shape Recognition Using Genetic Algorithm

International Conference on Evolutionary Computation, 1996

Shape recognition is a challenging task when shapesoverlap, forming noisy, occluded, partial shap... more Shape recognition is a challenging task when shapesoverlap, forming noisy, occluded, partial shapes.This paper uses a genetic algorithm for matchinginput shapes with model shapes described in termsof features such as line segments and angles (extractedusing traditional algorithms). The qualityof matching is gauged using a measure derived fromattributed shape grammars [12, 13]. Preliminary results,using shapes with about 30 features each, areextremely

Proceedings of the 2014 conference companion on Genetic and evolutionary computation companion - GECCO Comp '14, 2014

The mutation operator is the only genetic operator in Evolutionary Programming (EP). In the past ... more The mutation operator is the only genetic operator in Evolutionary Programming (EP). In the past researchers have nominated Gaussian, Cauchy, and Lévy distributions as mutation operators. According to the No Free Lunch theorem , no single mutation operator is able to outperform all others over the set of all possible functions. Potentially there is a lot of useful information generated when EP is ongoing. In this paper, we collect such information and propose a step size based self-adaptive mutation operator for Evolutionary Programming (SSEP). In SSEP, the mutation operator might be changed during the evolutionary process, based on the step size, from generation to generation. Principles for selecting an appropriate mutation operator for EP is proposed, with SSEP grounded on the principles. SSEP is shown to outperform static mutation operators in Evolutionary Programming on most of the functions tested. We also compare the experimental results of SSEP with other recent Evolutionary Programming methods, which uses multiple mutation operators.

There is a variety of approaches developed by researchers to solve different instances of timetab... more There is a variety of approaches developed by researchers to solve different instances of timetabling problems. During these studies different data formats are used to represent a timetabling problem instance and its solution, causing difficulties in the evaluation and comparison of approaches and sharing data. In this paper, a model for timetabling problems and a new XML data format for

Computer and Information Sciences III, 2012

Previous work on the Daily Car-Pooling problem includes an algorithm that consists of greedy assi... more Previous work on the Daily Car-Pooling problem includes an algorithm that consists of greedy assignment alternating with random perturbation. In this study, we examine the effect of varying the move acceptance policy, specifically Late-acceptance criteria with and without reheating. Late acceptance-based move acceptance criteria were chosen because there is strong empirical evidence in the literature indicating their superiority. Late-acceptance compares the objective values of the current solution with one which was obtained at a fixed number of steps prior to the current step during the search process in order to make an acceptance decision. We observe that the Late-acceptance criteria also achieve superior results in over 75% of cases for the Daily Car-Pooling problem, the majority of these results being statistically significant.

2012 12th UK Workshop on Computational Intelligence (UKCI), 2012

An iterative selection hyper-heuristic method controls and mixes a set of low-level heuristics wh... more An iterative selection hyper-heuristic method controls and mixes a set of low-level heuristics while solving a given problem. A low-level heuristic is selected and employed for improving a (set of) solution(s) at each step. This study investigates the influence of different heuristic selection methods within a population based incremental learning algorithm and hyper-heuristic based hybrid multiphase framework for solving dynamic environment problems. Even though the hybrid method delivers a good overall performance, it is superior in cyclic environments. The empirical results show that a heuristic selection method that relies on a fixed permutation of the underlying low-level heuristics, combined with a strategy that guarantees diversity when the environment changes is more successful than the learning approaches across different cyclic dynamic environments produced by a well known benchmark generator.

Proceedings of the 11th annual conference companion on Genetic and evolutionary computation conference - GECCO '09, 2009

If an optimisation algorithm performs a search in an environment that changes over time, it shoul... more If an optimisation algorithm performs a search in an environment that changes over time, it should be able to follow these changes and adapt itself for handling them in order to achieve good results. Different types of dynamics in a changing environment require the use of different approaches. Hyper-heuristics represent a class of methodologies that are high level heuristics performing search over a set of low level heuristics. Due to the generality of hyperheuristic frameworks, they are expected to be adaptive. Hence, a hyper-heuristic can be used in a dynamic environment to determine the approach to apply, adapting itself accordingly at each change. This study presents an initial investigation of hyper-heuristics in dynamic environments. A greedy hyper-heuristic is tested over a set of benchmark functions.

Lecture Notes in Computer Science, 2013

The vehicle routing problem (VRP) is a family of problems whereby a fleet of vehicles must servic... more The vehicle routing problem (VRP) is a family of problems whereby a fleet of vehicles must service the commodity demands of a set of geographically scattered customers from one or more depots, subject to a number of constraints. Early hyper-heuristic research focussed on selecting and applying a low-level heuristic at a given stage of an optimisation process. Recent trends have led to a number of approaches being developed to automatically generate heuristics for a number of combinatorial optimisation problems. Previous work on the VRP has shown that the application of hyper-heuristic approaches can yield successful results. In this paper we investigate the potential of grammatical evolution as a method to evolve the components of a variable neighbourhood search (VNS) framework. In particular two components are generated; constructive heuristics to create initial solutions and neighbourhood move operators to change the state of a given solution. The proposed method is tested on standard benchmark instances of two common VRP variants.

Lecture Notes in Computer Science, 2008

Genetic art is a recent art form generated by computers based on the genetic algorithms (GAs). In... more Genetic art is a recent art form generated by computers based on the genetic algorithms (GAs). In this paper, the components of a GA embedded into a genetic art tool named AMUSE are introduced. AMUSE is used to generate improvised melodies over a musical piece given a harmonic context. Population of melodies is evolved towards a better musical form based on a fitness function that evaluates ten different melodic and rhythmic features. Performance analysis of the GA based on a public evaluation shows that the objectives used by the fitness function are assembled properly and it is a successful artificial intelligence application.

Lecture Notes in Computer Science, 2007

Optimization in sports is a field of increasing interest. Combinatorial optimization techniques h... more Optimization in sports is a field of increasing interest. Combinatorial optimization techniques have been applied e.g. to game scheduling and playoff elimination. A common problem usually found in sports management is the assignment of referees to games already scheduled. There are a number of rules and objectives that should be taken into account when referees are assigned to games. We address a simplified version of a referee assignment problem common to many amateur leagues of sports such as soccer, baseball, and basketball. The problem is formulated by integer programming and its decision version is proved to be NP-complete. To tackle real-life large instances of the referee assignment problem, we propose a three-phase heuristic approach based on a constructive procedure, a repair heuristic to make solutions feasible, and a local search heuristic to improve feasible solutions. Numerical results on realistic instances are presented and discussed.

Information Sciences, 2015

Hyper-heuristics have emerged as automated high level search methodologies that manage a set of l... more Hyper-heuristics have emerged as automated high level search methodologies that manage a set of low level heuristics for solving computationally hard problems. A generic selection hyper-heuristic combines heuristic selection and move acceptance methods under an iterative single point-based search framework. At each step, the solution in hand is modified after applying a selected heuristic and a decision is made whether the new solution is accepted or not. In this study, we represent the trail of a hyper-heuristic as a third order tensor. Factorization of such a tensor reveals the latent relationships between the low level heuristics and the hyper-heuristic itself. The proposed learning approach partitions the set of low level heuristics into two equal subsets where heuristics in each subset are associated with a separate move acceptance method. Then a multi-stage hyper-heuristic is formed and while solving a given problem instance, heuristics are allowed to operate only in conjunction with the associated acceptance method at each stage. To the best of our knowledge, this is the first time tensor analysis of the space of heuristics is used as a data science approach to improve the performance of a hyper-heuristic in the prescribed manner. The empirical results across six different problem domains from a benchmark indeed indicate the success of the proposed approach.

Soft morphological filter optimization using a genetic algorithm for noise elimination

2014 14th UK Workshop on Computational Intelligence (UKCI), 2014

2014 IEEE Symposium on Evolving and Autonomous Learning Systems (EALS), 2014

Apprenticeship learning occurs via observations while an expert is in action. A hyper-heuristic i... more Apprenticeship learning occurs via observations while an expert is in action. A hyper-heuristic is a search method or a learning mechanism that controls a set of low level heuristics or combines different heuristic components to generate heuristics for solving a given computationally hard problem. In this study, we investigate into a novel apprenticeshiplearning-based approach which is used to automatically generate a hyper-heuristic for vehicle routing. This approach itself can be considered as a hyper-heuristic which operates in a train and test fashion. A state-of-the-art hyper-heuristic is chosen as an expert which is the winner of a previous hyper-heuristic competition. Trained on small vehicle routing instances, the learning approach yields various classifiers, each capturing different actions that the expert hyper-heuristic performs during the search process. Those classifiers are then used to produce a hyper-heuristic which is potentially capable of generalizing the actions of the expert hyperheuristic while solving the unseen instances. The experimental results on vehicle routing using the Hyper-heuristic Flexible (HyFlex) framework shows that the apprenticeship-learningbased hyper-heuristic delivers an outstanding performance when compared to the expert and some other previously proposed hyper-heuristics.

Multidisciplinary Scheduling: Theory and Applications, 2005

There is a variety of approaches developed by researchers to solve different instances of timetab... more There is a variety of approaches developed by researchers to solve different instances of timetabling problems. During these studies different data formats are used to represent a timetabling problem instance and its solution, causing difficulties in the evaluation and comparison of approaches and sharing data. In this paper, a model for timetabling problems and a new XML data format for them based on MathML is proposed.

A grouping hyper-heuristic framework: Application on graph colouring

Expert Systems with Applications, 2015

ABSTRACT Grouping problems are hard to solve combinatorial optimisation problems which require pa... more ABSTRACT Grouping problems are hard to solve combinatorial optimisation problems which require partitioning of objects into a minimum number of subsets while a given objective is simultaneously optimised. Selection hyper-heuristics are high level general purpose search methodologies that operate on a space formed by a set of low level heuristics rather than solutions. Most of the recently proposed selection hyper-heuristics are iterative and make use of two key methods which are employed successively; heuristic selection and move acceptance.

Detecting change and dealing with uncertainty in imperfect evolutionary environments

Information Sciences, 2015

ABSTRACT Imperfection of information is a part of our daily life; however, it is usually ignored ... more ABSTRACT Imperfection of information is a part of our daily life; however, it is usually ignored in learning based on evolutionary approaches. In this paper we develop an Imperfect Evolutionary System that provides an uncertain and chaotic imperfect environment that presents new challenges to its habitants. We then propose an intelligent methodology which is capable of learning in such environments. Detecting changes and adapting to the new environment is crucial to exploring the search space and exploiting any new opportunities that may arise. To deal with these uncertain and challenging environments, we propose a novel change detection strategy based on a Particle Swarm Optimization system which is hybridized with an Artificial Neural Network. This approach maintains a balance between exploitation and exploration during the search process. A comparison of approaches using different Particle Swarm Optimization algorithms show that the ability of our learning approach to detect changes and adapt as per the new demands of the environment is high.

International Journal of Parallel Programming, 2006

Determining the optimum data distribution, degree of parallelism and the communication structure ... more Determining the optimum data distribution, degree of parallelism and the communication structure on Distributed Memory machines for a given algorithm is not a straightforward task. Assuming that a parallel algorithm consists of consecutive stages, a Genetic Algorithm is proposed to find the best number of processors and the best data distribution method to be used for each stage of the parallel algorithm. Steady state genetic algorithm is compared with transgenerational genetic algorithm using different crossover operators. Performance is evaluated in terms of the total execution time of the program including communication and computation times. A computation intensive, a communication intensive and a mixed implementation are utilized in the experiments. The performance of GA provides satisfactory results for these illustrative examples.

A multi-level search framework for asynchronous cooperation of multiple hyper-heuristics

Proceedings of the 11th annual conference companion on Genetic and evolutionary computation conference - GECCO '09, 2009

ABSTRACT

Variable ordering has been a recurrent topic of study in the field of constraint satisfaction bec... more Variable ordering has been a recurrent topic of study in the field of constraint satisfaction because of its impact in the cost of the search. Various variable ordering heuristics have been proposed to help guiding the search under different situations. One important direction of the study about variable ordering is the use of distinct heuristics as the search progresses to reduce the cost of the search. Even though the idea of combining heuristics goes back to the 60's, only a few works that study which heuristics to use and how they interact with each other have been described. In this investigation, we analyse the interactions of four important variable ordering heuristics by combining them through hyper-heuristics that decide the heuristic to apply based on the depth of the nodes in the search tree. The paper does not include any specific model for generating such hyper-heuristics; instead, it presents an analysis of the changes in the cost when different heuristics are applied during the search by using one simple hyper-heuristic representation. The results show that selectively applying distinct heuristics as the search progresses may lead to important reductions in the cost of the search with respect to the performance of the same heuristics used in isolation.

A constraint satisfaction problem (CSP) is a combinatorial optimisation problem with many real wo... more A constraint satisfaction problem (CSP) is a combinatorial optimisation problem with many real world applications. One of the key aspects to consider when solving a CSP is the order in which the variables are selected to be instantiated. In this study, we describe a genetic programming hyper-heuristic approach to automatically produce heuristics for CSPs. Human-designed 'standard' heuristics are used as components enabling the construction of new variable ordering heuristics which is achieved through the proposed approach. We present empirical evidence that the heuristics produced by our approach are competitive considering the cost of the search when compared to the standard heuristics which are used to obtain the components for the new heuristics. The proposed approach is able to produce specialized heuristics for specific classes of instances that outperform the best standard heuristics for the same instances.

When solving a Constraint Satisfaction Problem (CSP), the order in which the variables are select... more When solving a Constraint Satisfaction Problem (CSP), the order in which the variables are selected to be instantiated has implications in the complexity of the search. This work presents the first ideas for evolving hyper-heuristics as decision matrices where the elements in the matrix represent the variable ordering heuristic to apply according to the constraint density and tightness of the current instance.