Search

The objective of this paper is to study how algorithms of optimization affect the parametersestimation of Autoregressive AR(1)Models. In our research we have represented the AR(1) models in linear state space form and applied the Kalman Filters to estimate the different unknown parameters of the model. Many methods have been proposed by researchers for the estimation of the parameters in the case of the linear state space models. In our work we have emphasized on the estimation through the Maximum Likelihood (ML). Statisticians have used many algorithms to optimise the likelihood function and they have proposed many filters; publishing their results in many papers. In spite of the fact that this field is so extended, we have emphasized our study in the financial field. Two quasi-Newton algorithms: Berndt, Hall, Hall, and Hausman (BHHH) and Broyden-Fletcher-Goldfarb-Shanno (BFGS), and the Expectation-Maximization (EM) algorithm have been chosen for this study. A practical study of these algorithms applied to the maximization of likelihood by means of the Kalman Filter have been done. The results are focused on efficiency in time of computation and precision of the unknown parameters estimation. A simulation study has been carried out, using as true values the parameters of this model published in the literature, in order to test the efficiency and precision of our implemented algorithms. [ABSTRACT FROM AUTHOR]

In this paper an algorithm for minimization of a nondifferentiable function is presented. The algorithm uses the Moreau-Yoshida regularization of the objective function and its second order Dini upper directional derivative. It is proved that the algorithm is well defined, as well as the convergence of the sequence of points generated by the algorithm to an optimal point. An estimate of the rate of convergence is given, too. [ABSTRACT FROM AUTHOR]

Since the evaluation of each individual is based on the time-consuming structural analysis, the computational efficiency of truss topology optimization using genetic algorithm is very low. The paper focuses on this challenging problem. It is observed that there are a number of duplicate individuals appearing repetitively in the evolutionary process. Therefore, an individual identification technique is introduced to avoid evaluating the duplicate individuals by the time-consuming structural analysis but by searching the evolutionary history data to save computing time, the computational complexity of this technique is deduced. The results of two truss examples verify that the technique can effectively improve the efficiency of the algorithm. Based on this identification technique, numeric experiments are implemented to study the influence of several factors, i.e., the population size, the max generation, and the scale of problems, on the proportion of duplicate individuals. Results show that the population size has a significant impact on the proportion, and that both the max generation and the scale of problems have little influence. [ABSTRACT FROM AUTHOR]

This paper discusses a new adaptive ant colony optimization algorithm and its characteristics are as follows: (1) a novel cranky ant who behaves strangely is introduced to prevent from trapping at the local optima, (2) a new observation technique for searching status is adopted to judge whether it is trapping at local optima. Experimental results using benchmark data prove that the proposed algorithm with the cranky ants and the observation technique enables to control the trade-off between intensification and diversification, in comparison with conventional ACO. [ABSTRACT FROM AUTHOR]

With the development of digital cameras, the digital photographs were flooded in our life. How to classify images efficiently in huge image database becomes an important research topic. In recent years, the related researches of the image classification are based on semantics. The scene image classification has received much attention especially because it contains plenty semantics. It is a difficult challenge to classify the scene images accurately. This paper tries to use particle swarm optimization (PSO) algorithm that has biological characteristic, and to train with the scene images of semantics. We can get a scene transformation matrix during the process. The scene transformation matrix can be used to classify scene images, which are close to human's semantics. The experiment shows our proposed method has great correct classification rate. [ABSTRACT FROM AUTHOR]

In this paper a new method for optimization of Elastic Bunch Graph Matching (EBGM) algorithm in frontal face recognition is presented. In EBGM algorithm, some pre-determined wavelength of Gabor wavelet is used to extract features from face image. For optimization of EBGM algorithm, Genetic Algorithm (GA) is used to select the best wavelengths of Gabor wavelet. For evaluation, algorithm has been tested on 300 classes of FERET face database. In training phase, only one image per class is trained. The recognition rate of optimized EBGM is about 91%. Also the optimized EBGM can run 1.5 times faster than original EBGM. [ABSTRACT FROM AUTHOR]

The adaptive immune algorithm (AIA) developed from immune algorithm (IA), owns faster computation speed and better convergence than that of GA and other stochastic type algorithms, due to its characteristic of having two layers optimization. The adaptive immune algorithm automatically adjusts the parameters to achieve fast convergence without falling in the local minimum point, according to the value of the distance between the antibodies. It leads to great reduction of the computation time, compared with other methods. The paper proposes to apply adaptive immune algorithm for reactive power optimization. The coding method for the control variables based on decimal system is introduced in detail. The test results of example systems demonstrate that the proposed reactive power optimization based on AIA method has advantages in terms of computation speed and convergence, and has great potential to be applied in practical power systems. [ABSTRACT FROM AUTHOR]

In this paper a new method based on the Ant-Miner algorithm is proposed to discover sets of unstructured classification rules. This method, called the Tree-Miner, creates a directed graph made up of nodes representing operators and operands. Each ant in a colony of artificial ants traverses this graph to find routes that represent the best unstructured rule antecedents. These antecedents are used to classify the given data and are also interpreted as knowledge hidden in the training data. The performance of the Tree-Miner algorithm was evaluated against that of the Ant-Miner according to the accuracy and the simplicity of the constructed rules. The results showed that our method has an acceptable predictive accuracy while discovering rules that are simpler and more comprehensive. [ABSTRACT FROM AUTHOR]

In this paper we present a new algorithm based on the Bee Colony Optimization (BCO) meta-heuristic for the Multidimensional Knapsack Problem (MKP), the goal of which is to find a subset of objects that maximizes a given objective function while satisfying some resource constraints. We show that our new algorithm obtains better results than Ant Colony Optimization algorithms and on most instances it reaches best known solutions. Especially we propose an efficient algorithm to produce randomly new solutions. [ABSTRACT FROM AUTHOR]

A hybrid population based Ant Colony Optimization (ACO) algorithm PFold-P-ACO for protein folding in the HP model is proposed in this paper. This is the first population based ACO algorithm in the bioinformatics. It is shown experimentally that the algorithms achieves on nearly all test sequences at least comparable results to other state of the art algorithms. Compared to the state of the art ACO algorithm PFold-P-ACO slightly better results and is faster on long sequences. [ABSTRACT FROM AUTHOR]

How to keep a balance between exploitation and exploration in Particle Swarm Optimization (PSO) for efficiently solving various optimization problems is an important issue. In order to handle premature convergence in PSO search, this paper proposes a novel algorithm, called Particle Swarm Optimization with Diversive Curiosity (PSO/DC), that introduces a mechanism of diversive curiosity into PSO for preventing premature convergence and ensuring exploration. A crucial idea here is to monitor the status of behaviors of swarm particles in PSO search by an interior indicator, and when swarm particles dropped into local minimum, they will be spontaneously reinitialized to start on finding other new solutions in search space. Applications of the proposal to a 2-dimensional optimization problem well demonstrate its effectiveness. Our experimental results indicate that the performance (90%) of the proposed method is superior in terms of success ratio to that (60%) of the PSO model optimized by EPSO. [ABSTRACT FROM AUTHOR]

Applications of the Particle Swarm Optimization (PSO) to solve image processing problem with a reference to a new automatic enhancement technique based on real-coded particle swarms is proposed in this paper. The enhancement process is a non-linear optimization problem with several constraints. The objective of the proposed PSO is to maximize an objective fitness criterion in order to enhance the contrast and detail in an image by adapting the parameters of a novel extension to a local enhancement technique. The feasibility of the proposed method is demonstrated and compared with Genetic Algorithms (GAs) based image enhancement technique. The obtained results indicate that the proposed PSO yields better results in terms of both the maximization of the number of pixels in the edges and the adopted objective evaluation. Computational time is also relatively small in the PSO case compared to the GA case. [ABSTRACT FROM AUTHOR]

In this paper, hybrid particle swarm optimization (PSO) is proposed for solving the challenging multi-contingency transient stability constrained optimal power flow (MC-TSCOPF) problem. The objective of this nonlinear optimization problem is to minimize the total fuel cost of the system and at the same time fulfil the transient stability requirements. The optimal power flow (OPF) with transient stability constraints considered is re-formulated as an extended OPF with additional rotor angle inequality constraints, which is suitable for hybrid PSO to solve. Comparison between various existing hybrid PSO techniques is carried out by solving the New England 39-bus system. Experimental results indicate that the hybrid PSO integrated with the mutation operation of genetic algorithms is better than the other existing hybrid PSO methods in both solution quality and stability. [ABSTRACT FROM AUTHOR]

A comparison of six basic Ant Colony Optimization (ACO) in dynamic environment was studied in this paper. Dynamic Traveling Salesman Problem (TSP) will be used as a dynamic environment. A number of cities are swap over time to make the TSP environment dynamic. A pheromone equalization strategy was applied in all the six ACO to react to the change. Three sets of TSP are used in this experiment. The result will show which of the six basic ant algorithms work best in dynamic environment. [ABSTRACT FROM AUTHOR]

The aim of 3-Parents Differential Evolution (3PDE) is to reduce the parental requirement in the original Differential Evolution (DE). The effectiveness of 3PDE has been reported and is considered as a useful algorithm that performed better convergence to optimality. In general, 3PDE is considered as a useful contribution since it has successfully reduced the parental requirement in DE without significant reductions in absolute optimization performance by gaining better average performance as well as stability. The objective of this paper is to investigate whether certain parameters that are self-adapted in 3PDE can enhance its performance for function optimization. Here, we propose three new algorithms to compare against 3PDE for their performance, which included 3SACr, 3SAF, and 3SAFCr. Fifty run were conducted for each 20 well-known benchmark functions to test all proposed algorithms. The experimental results showed that 3SACr performed the best among the other algorithms in terms of its better average performance as well stability. [ABSTRACT FROM AUTHOR]

Nature has always been the source of inspiration in the search of better heuristic methods. The Ant Colony Optimization(ACO) is also one of these kind of inspirations from the nature. The Ant Colony System(ACS) algorithm used for the solution of TSP is based on ACO. Here in this paper a new algorithm named ISTAR Ant Colony System, which is based on the ACS is proposed. [ABSTRACT FROM AUTHOR]

This paper presents a new metamodel-assisted metaheuristic algorithm for optimisation problems involving computationally expensive simulations. The algorithm, called Global Probing Search, is a population-based algorithm designed for global optimisation. The main idea of the algorithm is to introduce a probing phase in the creating of the new generation of the population. In this probing phase, a large number of candidate solutions are generated and a computationally cheap metamodel function is used for choosing the most promising candidates to transfer to the next generation. This approach could significantly enhance the efficiency of the optimisation process by avoiding wasting valuable evaluation time on solutions that are likely to be inferior. During the optimisation, the accuracy of the metamodel is constantly improved through on-line updating. The proposed algorithm is implemented on a real-world optimisation problem and initial results indicate that the algorithm show good performance in comparison with a standard Genetic Algorithm and an existing metamodel-assisted metaheuristic. [ABSTRACT FROM AUTHOR]

Particle Swarm Optimization (PSO) is a stochastic and population-based search algorithm that demonstrates its effctiveness in solving complex nonlinear optimization problems. Although the original PSO is very simple and effective, how to determine appropriate values of parameters in PSO is yet to be found. This paper proposes a novel method called evolutionary PSO, which estimates values of parameters in PSO for effectively finding globally optimal parameter values by a real-coded genetic algorithm. A crucial idea here is to adopt a temporary cumulative fitness instead of instantaneous fitness in a real-coded genetic algorithm for evaluating the performance of the PSO. It provides a useful measure that efficiently determines appropriate values of parameters in PSO. To demonstrate the effectiveness of the proposed method, we implement a simple computer experiment on a 2-dimensional optimization problem, and analyze the characteristics of dependency on initial condition. [ABSTRACT FROM AUTHOR]

Today numerous variants of heuristic optimization algorithms are used to solve different kinds of optimization problems. This huge variety makes it very difficult to reuse already implemented algorithms or problems. In this paper the authors describe a generic, extensible, and paradigm-independent optimization environment that strongly abstracts the process of heuristic optimization. By providing a well organized and strictly separated class structure and by introducing a generic operator concept for the interaction between algorithms and problems, HeuristicLab makes it possible to reuse an algorithm implementation for the attacking of lots of different kinds of problems and vice versa. Consequently HeuristicLab is very well suited for rapid prototyping of new algorithms and is also useful for educational support due to its state-of-the-art user interface, its self-explanatory API and the use of modern programming concepts. [ABSTRACT FROM AUTHOR]

Selection of an optimal subset of classifiers in designing classifier ensembles is an important problem. The search algorithms used for this purpose maximize an objective function which may be the combined training accuracy or diversity of the selected classifiers. Taking into account the fact that there is no benefit in using multiple copies of the same classifier, it is generally argued that the classifiers should be diverse and several measures of diversity are proposed for this purpose. In this paper, the relative strengths of combined training accuracy and diversity based approaches are investigated for the plurality voting based combination rule. Moreover, we propose a diversity measure where the difference in classification behavior exploited by the plurality voting combination rule is taken into account. [ABSTRACT FROM AUTHOR]

The paper proposes an implementation of the population learning algorithm designed to solve instances of the quadratic assignment problem. A short overview of the population- learning algorithm and a more detailed presentation of the proposed implementation is followed by the results of computational experiments carried. Particular attention is given to investigating performance characteristics and convergence of the PLA. Experiments have focused on identification of the probability distribution of solution time to a sub-optimal target value. [ABSTRACT FROM AUTHOR]

The algorithm about Brake's Initiative Time for Course Correction Fuze(CCF) has a large mount of calculation in the total trajectory calculation, so this part cost a majority of time when using DSP as processor. We should optimize the algorithm related to this part in order that the processor can accomplish the calculation for CCF quickly. This paper puts forward a kind of algorithm as approaching gradually to the real solution by a line of two points. Simulation has proved that this algorithm can reduce times of trajectory calculation and get the brake's initiative time more quickly. [ABSTRACT FROM AUTHOR]

This chapter has defined general optimization concepts including the idea that two types of optimization problems are associated with each six sigma project. One of these problems relates to the most desirable selection of methods to efficiently derive desirable settings. The chapter focuses on “stochastic optimization” with the solution methods using Monte Carlo for evaluating and comparing solutions. A simple genetic algorithm is provided useful for a wide variety of stochastic optimization problems including many robust system design and strategy design problems. [ABSTRACT FROM AUTHOR]