Showing papers on "Extremal optimization published in 2008"
TL;DR: This paper shows how ACO, which was initially developed to be a metaheuristic for combinatorial optimization, can be adapted to continuous optimization without any major conceptual change to its structure, and compares the results with those reported in the literature for other continuous optimization methods.
1,238 citations
13 May 2008
TL;DR: A bee colony optimization (BCO) algorithm for traveling salesman problem (TSP) is presented and experimental results comparing the proposed BCO model with some existing approaches on a set of benchmark problems are presented.
Abstract: A bee colony optimization (BCO) algorithm for traveling salesman problem (TSP) is presented in this paper. The BCO model is constructed algorithmically based on the collective intelligence shown in bee foraging behaviour. Experimental results comparing the proposed BCO model with some existing approaches on a set of benchmark problems are presented.
151 citations
TL;DR: A novel ant algorithm termed “continuous orthogonal ant colony” (COAC), whose pheromone deposit mechanisms would enable ants to search for solutions collaboratively and effectively and enhance the global search capability and accuracy.
Abstract: Research into ant colony algorithms for solving continuous optimization problems forms one of the most significant and promising areas in swarm computation. Although traditional ant algorithms are designed for combinatorial optimization, they have shown great potential in solving a wide range of optimization problems, including continuous optimization. Aimed at solving continuous problems effectively, this paper develops a novel ant algorithm termed “continuous orthogonal ant colony” (COAC), whose pheromone deposit mechanisms would enable ants to search for solutions collaboratively and effectively. By using the orthogonal design method, ants in the feasible domain can explore their chosen regions rapidly and efficiently. By implementing an “adaptive regional radius” method, the proposed algorithm can reduce the probability of being trapped in local optima and therefore enhance the global search capability and accuracy. An elitist strategy is also employed to reserve the most valuable points. The performance of the COAC is compared with two other ant algorithms for continuous optimization — API and CACO by testing seventeen functions in the continuous domain. The results demonstrate that the proposed COAC algorithm outperforms the others.
114 citations
TL;DR: This work proposes a novel hybrid approach specialized for the ATSP that incorporates an improved genetic algorithm (IGA) and some optimization strategies that contribute to its effectiveness.
97 citations
TL;DR: It turns out that for two of the four studied problems, the expected runtime for the considered class, expressed in terms of the problem size, is of the same order as that for (1+1)-Evolutionary Algorithm.
Abstract: The paper provides some theoretical results on the analysis of the expected time needed by a class of Ant Colony Optimization algorithms to solve combinatorial optimization problems. A part of the study refers to some general results on the expected runtime of the considered class of algorithms. These results are then specialized to the case of pseudo-Boolean functions. In particular, three well known functions and a combination of two of them are considered: the OneMax, the Needle-in-a-Haystack, the LeadingOnes, and the OneMax-Needle-in-a-Haystack. The results obtained for these functions are also compared to those from the well-investigated (1+1)-Evolutionary Algorithm. The results shed light on a suitable parameter choice for the considered class of algorithms. Furthermore, it turns out that for two of the four studied problems, the expected runtime for the considered class, expressed in terms of the problem size, is of the same order as that for (1+1)-Evolutionary Algorithm. For the other two problems, the results are significantly in favour of the considered class of Ant Colony Optimization algorithms.
80 citations
TL;DR: In order to extend EO to solve the multiobjective optimization problems, the Pareto dominance strategy is introduced to the fitness assignment of the proposed approach and a new novel elitist (1 + λ ) multi objective algorithm, called MOEO is proposed.
80 citations
TL;DR: It can be concluded that the extra time required by the ACO during the optimization process provides information that can be useful to deal with disturbances.
79 citations
01 Jan 2008
TL;DR: A new BPSO (NBPSO) is introduced and the results show the superiority of the INBPSO for solving optimization problems.
Abstract: Particle Swarm Optimization (PSO) algorithm, originated as a simulation of a simplified social system, is an evolutionary computation technique developed successfully in recent years and have been applied to many optimization problems. PSO can be applied to continuous and discrete optimization problems through local and global models. In this paper, PSO is addressed in details. There are some difficulties with the standard PSO where causing slow convergence rate on some optimization problems. These difficulties are transferred to the origin binary PSO (BPSO) that makes the algorithm not to converge well. Due to these difficulties with the BPSO, in this paper a new BPSO (NBPSO) is introduced. Several benchmark problems including unimodal and multimodal functions are considered for testing the robustness and effectiveness of the proposed method over the original BPSO. The results show that NBPSO performs much better than BPSO. Since the obtained results show that NBPSO may trap in the local optima, further modification is carried out. Two different methods are suggested to improve NBPSO which are denoted as Guaranteed Convergence BPSO (GCBPSO) and Improved NBPSO (INBPSO). The results show the superiority of the INBPSO for solving optimization problems.
78 citations
TL;DR: In this article, a hybrid evolutionary algorithm with marriage of GA and extremal optimization (EO) was proposed for solving a class of production scheduling problems in manufacturing, which is characterized by two major requirements: (i) selecting a subset of orders from manufacturing orders to be processed; (ii) determining the optimal production sequence under multiple constraints.
Abstract: This paper presents a hybrid evolutionary algorithm with marriage of genetic algorithm (GA) and extremal optimization (EO) for solving a class of production scheduling problems in manufacturing The scheduling problem, which is derived from hot rolling production in steel industry, is characterized by two major requirements: (i) selecting a subset of orders from manufacturing orders to be processed; (ii) determining the optimal production sequence under multiple constraints, such as sequence-dependant transition costs, non-execution penalties, earliness/tardiness (E/T) penalties, etc A combinatorial optimization model is proposed to formulate it mathematically For its NP-hard complexity, an effective hybrid evolutionary algorithm is developed to solve the scheduling problem through combining the population-based search capacity of GA and the fine-grained local search efficacy of EO The experimental results with production scale data demonstrate that the proposed hybrid evolutionary algorithm can provide superior performances in scheduling quality and computation efficiency
71 citations
01 Jan 2008
TL;DR: This chapter gives an overview over approximation methods in multi-objective combinatorial optimization, and focuses on recent approaches, where metaheuristics are hybridized and/or combined with exact methods.
Abstract: Many real-world optimization problems can be modelled as combinatorial optimization problems Often, these problems are characterized by their large size and the presence of multiple, conflicting objectives Despite progress in solving multi-objective combinatorial optimization problems exactly, the large size often means that heuristics are required for their solution in acceptable time Since the middle of the nineties the trend is towards heuristics that “pick and choose” elements from several of the established metaheuristic schemes Such hybrid approximation techniques may even combine exact and heuristic approaches In this chapter we give an overview over approximation methods in multi-objective combinatorial optimization We briefly summarize “classical” metaheuristics and focus on recent approaches, where metaheuristics are hybridized and/or combined with exact methods
22 Sep 2008
TL;DR: This paper studies this combination of ACO and local search from a theoretical point of view and point out situations where introducing local search into an ACO algorithm enhances the optimization process significantly, and illustrates the drawback that such a combination might have by showing that this may prevent an ACOs from obtaining optimal solutions.
Abstract: Ant colony optimization (ACO) is a metaheuristic that produces good results for a wide range of combinatorial optimization problems. Often such successful applications use a combination of ACO and local search procedures that improve the solutions constructed by the ants. In this paper, we study this combination from a theoretical point of view and point out situations where introducing local search into an ACO algorithm enhances the optimization process significantly. On the other hand, we illustrate the drawback that such a combination might have by showing that this may prevent an ACO algorithm from obtaining optimal solutions.
20 Oct 2008
TL;DR: An improved ant colony optimization algorithm for traveling salesman problem is presented, which adopts a new probability selection mechanism by using Held-Karp lower bound to determine the trade-off between the influence of the heuristic information and the pheromone trail.
Abstract: The traveling salesman problem (TSP) in operations research is a classical problem in discrete or combinatorial optimization. It is a prominent illustration of a class of problems in computational complexity theory which are classified as NP-hard. Ant colony optimization inspired by co-operative food retrieval have been widely applied unexpectedly successful in the combinatorial optimization. This paper presents an improved ant colony optimization algorithm for traveling salesman problem, which adopts a new probability selection mechanism by using Held-Karp lower bound to determine the trade-off between the influence of the heuristic information and the pheromone trail. The experiments showed that it can stably generate better solution for the traveling salesman problem than rank-based ant system and max-min ant colony optimization algorithm.
TL;DR: A novel Pareto-based algorithm, which can be regarded as an extension of EO, to solve multiobjective optimization problems, is presented and demonstrated that MOPEO is competitive with the state-of-the-art multiobjectives evolutionary algorithms.
Abstract: In recent years, a general-purpose local-search heuristic method called Extremal Optimization (EO) has been successfully applied in some NP-hard combinatorial optimization problems In this paper, we present a novel Pareto-based algorithm, which can be regarded as an extension of EO, to solve multiobjective optimization problems The proposed method, called Multiobjective Population-based Extremal Optimization (MOPEO), is validated by using five benchmark functions and metrics taken from the standard literature on multiobjective evolutionary optimization The experimental results demonstrate that MOPEO is competitive with the state-of-the-art multiobjective evolutionary algorithms Thus MOPEO can be considered as a viable alternative to solve multiobjective optimization problems
01 Jun 2008
TL;DR: A new model based on particle swarm optimization to detect network community is proposed and experimental results indicate this model can effectively find web communities of network structure without any domain information.
Abstract: Web community detection is one of the important ways to enhance retrieval quality of web search engine. How to design one highly effective algorithm to partition network community with few domain knowledge is the key to network community detection. Traditional algorithms, such as Wu-Huberman algorithm, need priori information to detect community, the Radichi algorithm relies on the triangle number in the network, the extremal optimization algorithm proposed by Duch J. is extremely sensitive to the initial solution, easy to fall into the local optimum. This article proposes a new model based on particle swarm optimization to detect network community, and with different scale network chart, Zachary, Krebs and dolphins network architecture to test the algorithm, the experimental results indicate this model can effectively find web communities of network structure without any domain information.
19 Dec 2008
TL;DR: Details of the algorithm used to solve MJQO problem have been interpreted, including how to define heuristic information, how to implement local pheromone update and global pheroma update and how to design state transition rule and the simulation result indicates that ACO is more effective and efficient.
Abstract: Multi-join query optimization (MJQO) is an important technique for designing and implementing database manage system. It is a crucial factor that affects the capability of database. This paper proposes a new algorithm to solve the problem of MJQO based on ant colony optimization (ACO). In this paper, details of the algorithm used to solve MJQO problem have been interpreted, including how to define heuristic information, how to implement local pheromone update and global pheromone update and how to design state transition rule. After repeated iteration, a reasonable solution is obtained. Compared with genetic algorithm, the simulation result indicates that ACO is more effective and efficient.
Proceedings Article•
06 Feb 2008TL;DR: Simulation results show that the proposed TACO algorithm outperforms existing neural network based approaches in solution quality and demonstrates the feasibility of the proposed approach in multi-robot path planning.
Abstract: In this paper, a Team ant colony optimization algorithm (TACO) is proposed for the multiple travelling salesman problem with MinMax objective. The novel idea is to replace every ant in an ant colony optimization algorithm, for example Ant Colony System [1], with a team of ants and letting those teams construct solutions to the multiple travelling salesman problem. The simulation results show that the proposed algorithm outperforms existing neural network based approaches in solution quality. Furthermore, the presented experiments demonstrate the feasibility of the proposed approach in multi-robot path planning.
15 Sep 2008
TL;DR: This paper proposes a discrete binary version of the electromagnetism-like method for solving the combinatorial optimization problems and concludes that the method is capable of solving such well-known problems more efficiently than the previous works.
Abstract: The electromagnetism-like method (EM) is a meta-heuristic algorithm utilizing an attraction-repulsion mechanism to move sample points (i.e., our solutions) towards the optimality. In general, the EM method has been initially used for solving continuous optimization problems and could not be applied on combinatorial optimization ones. This paper proposes a discrete binary version of the electromagnetism-like method (EM) for solving the combinatorial optimization problems. To show the efficiency of our proposed EM, we use it for solving the traveling salesman problem and compare our computational results with those reported in the literature. Finally we conclude that our method is capable of solving such well-known problems more efficiently than the previous works.
TL;DR: This chapter discusses the application of Opposition-Based Optimization (OBO) to ant algorithms and aims to improve the accuracy and convergence of the current algorithm by extending it with the concept of OBO.
Abstract: The chapter discusses the application of Opposition-Based Optimization (OBO) to ant algorithms. Ant Colony Optimization (ACO) is a powerful optimization technique that has been used to solve many complex problems. Despite its successes, ACO is not a perfect algorithm: it can remain trapped in local optima, miss a portion of the solution space or, in some cases, it can be slow to converge. Thus, we were motivated to improve the accuracy and convergence of the current algorithm by extending it with the concept of OBO. In the case of ACO, the application of opposition can be challenging because ACO usually optimizes using a graph representation of problems, where the opposite of solutions and partial components of the solutions are not clearly defined.
Journal Article•
TL;DR: The basic version of DE and its modifications are presented, and their advantages and disadvantages are discussed and some issues for further research on DE are addressed.
Abstract: Differential evolution(DE)is a heuristic global optimization technique based on population.It is robust for real parameter optimization.To speed up the optimization and overcome the premature convergence of the heuristic optimization technique,many modifications are made to DE.The basic version of DE and its modifications are presented,and their advantages and disadvantages are also discussed.Some issues for further research on DE are addressed.
TL;DR: Some current research within the ACO community is outlined, reporting recent results obtained on different problems, and a particular research line is focused on, named ANTS, providing some details on the algorithm and presenting results recently obtained on a prototypical strongly constrained problem: the set partitioning problem.
Abstract: Ant Colony Optimization (ACO) is a class of metaheuristic algorithms sharing the common approach of constructing a solution on the basis of information provided both by a standard constructive heuristic and by previously constructed solutions. This article is composed of three parts. The first one frames ACO in current trends of research on metaheuristics for combinatorial optimization. The second outlines some current research within the ACO community, reporting recent results obtained on different problems, while the third part focuses on a particular research line, named ANTS, providing some details on the algorithm and presenting results recently obtained on a prototypical strongly constrained problem: the set partitioning problem.
12 Jul 2008
TL;DR: This paper covers a multi-objective Ant Colony Optimization, which is applied to the NP-complete multi-Objective shortest path problem in order to approximate Pareto-fronts.
Abstract: This paper covers a multi-objective Ant Colony Optimization, which is applied to the NP-complete multi-objective shortest path problem in order to approximate Pareto-fronts. The efficient single-objective solvability of the problem is used to improve the results of the ant algorithm significantly. A dynamic program is developed which generates local heuristic values on the edges of the problem graph. These heuristic values are used by the artificial ants.
TL;DR: This paper investigates the capabilities of the ant colony optimization (ACO) heuristic for solving the Max-cut problem and presents an AntCut algorithm that can solve the problem more efficiently and effectively.
Abstract: Max-cut problem is an NP-complete and classical combinatorial optimization problem that has a wide range of applications in different domains, such as bioinformatics, network optimization, statistical physics, and very large scale integration design. In this paper we investigate the capabilities of the ant colony optimization (ACO) heuristic for solving the Max-cut problem and present an AntCut algorithm. A large number of simulation experiments show that the algorithm can solve the Max-cut problem more efficiently and effectively.
01 Jun 2008
TL;DR: An ACO approach to optimal control is proposed, which requires that a continuous-time, continuous-state model of the system, together with a finite action set, is formulated as a discrete, non-deterministic automaton.
Abstract: Ant Colony Optimization (ACO) has proven to be a very powerful optimization heuristic for Combinatorial Optimization Problems (COPs). It has been demonstrated to work well when applied to various NP-complete problems, such as the traveling salesman problem. In this paper, an ACO approach to optimal control is proposed. This approach requires that a continuous-time, continuous-state model of the system, together with a finite action set, is formulated as a discrete, non-deterministic automaton. The control problem is then translated into a stochastic COP. This method is applied to the time-optimal swing-up and stabilization of a pendulum.
18 Oct 2008
TL;DR: To obtain a more reasonable solution, a mergence mechanism, based on the local search result of each ant, is employed and the proposed method has better performance than the conventional ACO algorithm.
Abstract: An improved ant colony optimization algorithm is proposed in this paper. Comparing with the conventional ant colony optimization algorithm, the proposed method has two highlights. First, a newly strategy based on the dynamic control of solution construction is adopted. The purpose of this strategy is to ensure ants to exploit the solutions at the beginning of searching procedure with large probability while at the end of the searching procedure the solutions provided by each ant are obtained by searching around the best-so-far solution. Second, to obtain a more reasonable solution, a mergence mechanism, based on the local search result of each ant, is employed. The experiments demonstrate that the proposed method has better performance than the conventional ACO algorithm.
15 Aug 2008
TL;DR: Based on classical ant algorithm, a method for solving optimization problem with continuous parameters using ant colony algorithm is proposed, which has much higher convergence speed and the disadvantage of classical ant colony algorithms of not being suitable for solving continuous optimization problems is overcome.
Abstract: One of the most promising innovations in the area of heuristics is the development of evolutionary algorithms. A valuable and novel proposition in this area is ant algorithms. Researchers examining the behavior of real ants developed algorithms and applied them to many optimization problems. Based on classical ant algorithm, a method for solving optimization problem with continuous parameters using ant colony algorithm is proposed in this paper. In the method, the size of artificial ant colony is determined according to the constrained field of the problem, the amount of the change in objective function is introduced as heuristic factor of the algorithm. The searching region is reduced, moved and modified according to the transition probability dynamically. Our experimental results in continuous optimization problem show that this method has much higher convergence speed and the disadvantage of classical ant colony algorithm of not being suitable for solving continuous optimization problems is overcome.
25 Sep 2008
TL;DR: This model can be used as a meta-model for heuristic optimization algorithms, enabling users to represent custom algorithms in a flexible way by still providing a broad spectrum of reusable algorithm building blocks.
Abstract: The definition of a generic algorithm model for representing arbitrary heuristic optimization algorithms is one of the most challenging tasks when developing heuristic optimization software systems. As a high degree of flexibility and a large amount of reusable code are requirements that are hard to fulfill together, existing frameworks often lack of either of them to a certain extent. To overcome these difficulties the authors present a generic algorithm model not only capable of representing heuristic optimization but that can be used for modeling arbitrary algorithms. This model can be used as a meta-model for heuristic optimization algorithms, enabling users to represent custom algorithms in a flexible way by still providing a broad spectrum of reusable algorithm building blocks.
18 Oct 2008
TL;DR: A novel ant colony optimization solving method based on the time window constraints for the prize-collecting traveling salesman problem with time windows is presented.
Abstract: Focused on a variation of the Euclidean traveling salesman problem (TSP), namely the prize-collecting traveling salesman problem with time windows (PCTSPTW), this paper presents a novel ant colony optimization solving method. The time window constraints are considered in the computation for the probability of selection of the next city. The parameters of the algorithm are analyzed by experiments. Numerical results also show that the proposed method is effective for the PCTSPTW problem.
TL;DR: A combinatorial optimization algorithm is adapted for the search problem in computational protein design that takes advantage of the knowledge of local energy information and systematically improves on the residues that have high local energies.
Abstract: We adapt a combinatorial optimization algorithm, extremal optimization (EO), for the search problem in computational protein design. This algorithm takes advantage of the knowledge of local energy information and systematically improves on the residues that have high local energies. Power-law probability distributions are used to select the backbone sites to be improved on and the rotamer choices to be changed to. We compare this method with simulated annealing (SA) and motivate and present an improved method, which we call reference energy extremal optimization (REEO). REEO uses reference energies to convert a problem with a structured local-energy profile to one with more random profile, and extremal optimization proves to be extremely efficient for the latter problem. We show in detail the large improvement we have achieved using REEO as compared to simulated annealing and discuss a number of other heuristics we have attempted to date. © 2008 Wiley Periodicals, Inc. J Comput Chem, 2008
25 Sep 2008
TL;DR: This paper wants to demonstrate how the general and open optimization environment HeuristicLab in its latest version can be used to optimize simulation models.
Abstract: Simulation optimization today is an important branch in the field of heuristic optimization problems. Several simulators include built-in optimization and several companies have emerged that offer optimization strategies for different simulators. Often the optimization strategy is a secret and only sparse information is known about its inner workings. In this paper we want to demonstrate how the general and open optimization environment HeuristicLab in its latest version can be used to optimize simulation models.