Neural networks are usually trained using local gradient-based procedures. Such methods frequently found sub-optimal solutions being trapped in local minima. In solving some application problems, the input/output data...
详细信息
Neural networks are usually trained using local gradient-based procedures. Such methods frequently found sub-optimal solutions being trapped in local minima. In solving some application problems, the input/output data sets used to train a neural network may not be hundred percent precise but within certain range. It is difficult for the traditional neural network to solve such problems. A learning algorithm based on interval optimization is presented in this paper. The above disadvantages of the traditional learning algorithm are settled by using this method.
Non-functional requirements/characteristics are important for providing effectively every kind of services including web services. A realistic web service must meet both functional and non-functional requirements of i...
详细信息
ISBN:
(纸本)1595934359
Non-functional requirements/characteristics are important for providing effectively every kind of services including web services. A realistic web service must meet both functional and non-functional requirements of its consumers. Therefore, it is important that a web services framework is augmented so that non-functional characteristics of a web service can be determined at run-time and consumers are bound to a service that best meet their functional as well as non-functional requirements. In this paper, we propose an extension of the existing web services framework that enables a collection of functional and non-functional service characteristics at run-time, and usage of the collected data in discovery, binding, and execution of web services. Descriptions of new and enhanced components of the proposed framework are also presented. Typical publishing and usage scenarios in the proposed framework are also described.
Visualization in scientific computing and engineering design is receiving wide attention. It can assist engineers, scientists, and technicians to access, analyze, manage, visualize, and present large and diverse quant...
详细信息
Frequent mine gas incident in china fully shows there must be many undiscovered essential factors in all previous gas incident investigations. Exploring a new analysis approach is vital to prevent and control mine gas...
详细信息
Integrated virtual reality with Intelligent Tutoring System, a multi-agent architecture was proposed for intelligent virtual training system (IVTS) for mine safety training. In order to make sure IVTS agent's task...
详细信息
Swarm-diversity is an important factor influencing the global convergence of Particle Swarm Optimization (PSO). In order to overcome the premature convergence, the paper develops a diversity-controlled PSO (DCPSO). DC...
详细信息
Swarm-diversity is an important factor influencing the global convergence of Particle Swarm Optimization (PSO). In order to overcome the premature convergence, the paper develops a diversity-controlled PSO (DCPSO). DCPSO adopts swarm-diversity as a dynamic indicator to control the tuning of the parameters in a search, which in turn, can not only modify the swarm diversity, but also manipulate the exploitation and the exploration adaptively. Some experiments have been done and the results show DCPSO performs very well on benchmark optimization problems, and outperforms the basic PSO with a robust global convergent ability.
Moving object segmentation in the compressed domain plays an important role in many real-time applications, e.g. video indexing, video surveillance, etc. H.264/AVC is the up-to-date video coding standard, which employ...
详细信息
ISBN:
(纸本)300018726X
Moving object segmentation in the compressed domain plays an important role in many real-time applications, e.g. video indexing, video surveillance, etc. H.264/AVC is the up-to-date video coding standard, which employs several new coding tools and provides a different video format. In this paper, a robust spatio-temporal segmentation approach to extract moving objects in the compressed domain for H.264 video stream is proposed. Coarse moving objects are firstly extracted from the motion vector field through the MRF classification process. Spatial segmentation is based on the HT coefficient, which is used to refine the object boundary of temporal segmentation results. Fusion of spatio-temporal segmentation is based on region projection based labeling technique. Further edge refinement is performed to smooth the object boundaries. The performance of the proposed approach is illustrated by simulation carried on several test sequences. Experimental results show that the proposed approach can provide the visually similar results as by using spatial domain process with much less computational complexity.
In order to improve the global convergent ability of the standard particle swarm optimization (SPSO), the paper develops a new version of particle swarm optimization guided by the acceleration information (AGPSO). Fir...
详细信息
ISBN:
(纸本)1424406048
In order to improve the global convergent ability of the standard particle swarm optimization (SPSO), the paper develops a new version of particle swarm optimization guided by the acceleration information (AGPSO). Firstly, the paper introduces the concept of acceleration into the AGPSO version and makes a convergent analysis of the new model. Secondly, the paper studies the parameter choices of the AGPSO model. Thirdly, the paper provides the A GPSO with an oscillating factor to adjust the influence of the acceleration on the velocity, which can guarantee the AGPSO to converge to the global optimization validly. Finally, the proposed AGPSO versions are used to some benchmark optimizations, the experimental results show those AGPSO versions can overcome the premature problem validly, and outperforms the standard PSO in the global search ability with a quicker convergent speed
This paper analyzes the requirement of authorization service for grid computing systems and proposes the use of threshold closure as a basic mechanism for implementing authorization service in grid computing systems. ...
详细信息
This paper analyzes the requirement of authorization service for grid computing systems and proposes the use of threshold closure as a basic mechanism for implementing authorization service in grid computing systems. While pointing out the desirable features of threshold closure for complex authorization policies, the paper also discusses the practical limitations of threshold closure in such an environment, and then puts forward a new authorization service for virtual organization. In addition, an access control protocol which is based on PKI is designed in the paper. By segregating the policy and mechanism aspects of threshold closure, the new service can use existing security infrastructure in grid computing system while keep the ability to express complex authorization policy effectively
Becausemining complete set of frequent patterns from dense database could be impractical, an interesting alternative has been proposed recently. Instead of mining the complete set of frequent patterns, the new model o...
详细信息
Becausemining complete set of frequent patterns from dense database could be impractical, an interesting alternative has been proposed recently. Instead of mining the complete set of frequent patterns, the new model only finds out the maximal frequent patterns, which can generate all frequent patterns. FP-growth algorithm is one of the most efficient frequent-pattern mining methods published so far. However,because FP-tree and conditional FP-trees must be two-way traversable, a great deal memory is needed in process of mining. This paper proposes an efficient algorithm Unid_FP-Max for mining maximal frequent patterns based on unidirectional FP-tree. Because of generation method of unidirectional FP-tree and conditional unidirectional FP-trees, the algorithm reduces the space consumption to the fullest extent. With the development of two techniques:single path pruning and header table pruning which can cut down many conditional unidirectional FP-trees generated recursively in mining process, Unid_ FP-Max further lowers the expense of time and space.
暂无评论