This paper presents a canonical duality theory for solving quadratic minimization problems subjected to either box or integer constraints. Results show that under Gao and Strang's general global optimality conditi...
详细信息
This paper presents a canonical duality theory for solving quadratic minimization problems subjected to either box or integer constraints. Results show that under Gao and Strang's general global optimality condition, these well-known nonconvex and discrete problems can be converted into smooth concave maximization dual problems over closed convex feasible spaces without duality gap, and can be solved by well-developed optimization methods. Both existence and uniqueness of these canonical dual solutions are presented. Based on a second-order canonical dual perturbation, the discrete integer programming problem is equivalent to a continuous unconstrained Lipschitzian optimization problem, which can be solved by certain deterministic technique. Particularly, an analytical solution is obtained under certain condition. A fourth-order canonical dual perturbation algorithm is presented andapplications are illustrated. Finally, implication of the canonical duality theory for the popular semi-definite programming method is revealed.
In this paper we present an extension of the Boolean Neural Networks BNN [1] to solve a quadratic 0-1 programming problem under linear constraints. The network is called a QBNN (Quadratic Boolean Neural Network). To d...
详细信息
Current trends in Big Data processing indicate that the volume, velocity and variety of data are increasing quickly due to an explosion on diversity and number of sources of information. This poses challenges for Big ...
详细信息
We investigate the use of partial functions, fitness sharing and committee learning in genetic programming. The primary intended application of the work is in learning spatial relationships for ecological modelling. T...
详细信息
We investigate the use of partial functions, fitness sharing and committee learning in genetic programming. The primary intended application of the work is in learning spatial relationships for ecological modelling. The approaches are evaluated using a well-studied ecological modelling problem, the greater glider population density problem. Combinations of the three treatments (partial functions, fitness sharing and committee learning) are compared on the dimensions of accuracy and computational cost. Fitness sharing significantly improves learning accuracy, and populations of partial functions substantially reduce computational cost. The results of committee learning are more equivocal, and require further investigation. The learned models are highly predictive, but also highly explanatory. (C) 2001 Elsevier Science BN. All rights reserved.
The widespread deployment of networked applications and adoption of the internet has fostered an environment in which many distributed services are available. There is great demand to automate business processes and w...
详细信息
The widespread deployment of networked applications and adoption of the internet has fostered an environment in which many distributed services are available. There is great demand to automate business processes and workflows among organizations and individuals. Solutions to such problems require orchestration of concurrent and distributed services in the face of arbitrary delays and failures of components and communication. We propose a novel approach, called Orc for orchestration, that supports a structured model of concurrent and distributed programming. This model assumes that basic services, like sequential computation and data manipulation, are implemented by primitive sites. Orc provides constructs to orchestrate the concurrent invocation of sites to achieve a goal - while managing time-outs, priorities, and failure of sites or communication.
Cloud computing can describe on-demand administration and prerequisite of resources, and information as services on the cloud. The dynamic environment of the cloud causes various unexpected obstacles. The ability of a...
详细信息
ISBN:
(纸本)9781509067107
Cloud computing can describe on-demand administration and prerequisite of resources, and information as services on the cloud. The dynamic environment of the cloud causes various unexpected obstacles. The ability of a system to respond normally to unexpected equipment andprogramming fault is called fault tolerance. In order to realize robustness and reliability in cloud computing, it is necessary to evaluate and process obstacles effectively. Various fault detection methods and architecture models have been proposed to increase the fault tolerance capability of the cloud. Consider the fact, this paper has been paying attention on solving the problem of faults in the cloud computing. This research has conducted to outline the work that has been done in this field which represents the basic concept of fault tolerance developed by different researchers and various algorithms used to resolve the fault tolerance problem in cloud computing.
Soft computing is a new approach to computing. It has ability to reason and learn in an environment of uncertainty, approximation and imprecision. Soft computing combines many technologies like fuzzy logic, probabilis...
详细信息
Multi-thread is a mature and widely-used programming mode for multitasking applications developing, especially in Java. However *** is not designed for HPC parallel programming. Java multi-thread program is confined w...
详细信息
Classification of signals acquired by condition monitoring systems for automotive application is becoming increasingly important. The work presented in this paper is motivated by a real-life classification challenge o...
详细信息
ISBN:
(纸本)9781479983490
Classification of signals acquired by condition monitoring systems for automotive application is becoming increasingly important. The work presented in this paper is motivated by a real-life classification challenge organized by Ford. Data samples from an automotive subsystem were collected. A classifier is designed to robustly isolate the different types of problems, by analyzing the acquired signals. In this paper, the wavelet transform is used as data reduction and feature selection tool. The proper input feature is then classified by neural network as a nonlinear classifier tool. Results show significant accuracy with reduced amount of false positives.
This paper is on how to build an investment model of security investment fund which will produce the maximum in profits. The deficiencies in Markowitz's portfolio selection decision model are analysed. In this pap...
详细信息
ISBN:
(纸本)9780769534022
This paper is on how to build an investment model of security investment fund which will produce the maximum in profits. The deficiencies in Markowitz's portfolio selection decision model are analysed. In this paper, the writers use nonlinear and dynamic model depending on Return-Variance model under the limited condition to decide how to invest. The design method of fund portfolio is presented which can optimize portfolio gain with the mixture experience design applying.
暂无评论