On the basis of fuzzy optimization model,the concept of equivalent error function is introduced to establish a new back propagation(bp) algorithm of Hessian matrix to promote the *** Newton's iteration method is a...
详细信息
On the basis of fuzzy optimization model,the concept of equivalent error function is introduced to establish a new back propagation(bp) algorithm of Hessian matrix to promote the *** Newton's iteration method is applied to promote the training efficiency and accelerate the *** method is used in the decision-making analysis for economic,water resources and environmental planning of the Dalian City.
On the basis of fuzzy optimization model, the concept of equivalent error function is introduced to establish a new back propagation (bp) algorithm of Hessian matrix to promote the calculation. The New...
详细信息
On the basis of fuzzy optimization model, the concept of equivalent error function is introduced to establish a new back propagation (bp) algorithm of Hessian matrix to promote the calculation. The Newton's iteration method is applied to promote the training efficiency and accelerate the convergence. This method is used in the decision-making analysis for economic, water resources and environmental planning of the Dalian City.
This paper proposes an automatic detection of oil spills on SAR （synthetic aperture radar） images using DE （differential evolution）, neutral network and bp （back propagation） algorithm. Here, DE and bp are combi...
详细信息
This paper proposes an automatic detection of oil spills on SAR （synthetic aperture radar） images using DE （differential evolution）, neutral network and bp （back propagation） algorithm. Here, DE and bp are combined to train a multilayer perceptron （MLP） network for achieving the global extreme with a better convergence speed. The input data of neural networks are the geometrical characteristics ofoil spills （e.g. area, perimeter, complexity） and the physical behavior ofoil spills （e,g. mean or max backscatter value, standard deviation of the dark formation）. The out data are oil spill or look-alike. We experiment ALOS/PALSAR and EnviSAT ASAR on East sea area of Viet Nam. The experimental results show that the combination algorithm converges faster and has significantly better capability of avoiding local optima.
On-Line Learning Behavior depend on learning subject self-control learning, collaborative learning, and obtaining of support and help. Based on learning subject, the On-Line learning behavior need the real-time monito...
详细信息
ISBN:
(纸本)9781467383028
On-Line Learning Behavior depend on learning subject self-control learning, collaborative learning, and obtaining of support and help. Based on learning subject, the On-Line learning behavior need the real-time monitoring and the effective instruction to through the evaluation in the learning process. The bp algorithm model of evaluating ELearning behavior selects the learning behavior which affects the study effect in network learning process, has established the network learning behavior evaluating indicator system, and takes the data-in by the second-level target, carries on the network training using MATLAB, In the network training process, the global error assumes the declining trend basically, the restraining effect is good. Through the test indicated that this model may use in evaluating the network learning behavior, and obtains the expectation effect.
Standing at the perspective of the project management, this thesis, from the relation between schedule, quality and cost, analyzes how the project cost was influenced by the schedule risk and quality risk, assesses th...
详细信息
Standing at the perspective of the project management, this thesis, from the relation between schedule, quality and cost, analyzes how the project cost was influenced by the schedule risk and quality risk, assesses the total cost risk of the project with artificial neural network bp algorithm, and determines the most sensitive factors of the project cost risk, which could provide reference for the project managers to control the project cost risk.
This paper presents a bp network based on improved particle swarm optimization to solve the selective harmonic elimination technique of switch angles. One of the difficulties of selective harmonic elimination techniqu...
详细信息
ISBN:
(纸本)9781467374439
This paper presents a bp network based on improved particle swarm optimization to solve the selective harmonic elimination technique of switch angles. One of the difficulties of selective harmonic elimination technique is solving the switch angles, the traditional method has shortcomings such as initial value selection is difficult and the iterative process complex. Traditional bp algorithm also has shortcomings such as slow convergence speed and easy to fall into local weights. Use the improved particle swarm algorithm to optimize neural network's weights, threshold and connection structure. Then use the trained network to solve SHEPWM switching angles. The simulation results show that the optimized bp network convergence rate increased significantly, and solving the switch angles accuracy is improved significantly.
For the purpose of this article,there is a huge amount of work in feature extraction,selection and other aspects using traditional supervised machine learning methods,and features for different applications often requ...
详细信息
ISBN:
(纸本)9781510823808
For the purpose of this article,there is a huge amount of work in feature extraction,selection and other aspects using traditional supervised machine learning methods,and features for different applications often required different scenarios which need to manually design,but with the final result not *** paper shows a unsupervised feature extraction method-combine Stacked Auto Encoder and Support Vector Machine,experiments had shown that the algorithm's accuracy is 99.31% in MINIST better than other *** study can help Handwritten Digits Recognition get better development in various fields,such as ZIP code automatic identification,automatic processing of bank checks.
Low Density Parity Check(LDPC) code itself has a good performance and the application of LDPC in the communication field is extensive. LDPC codes have the characteristics of low bit error rate, easy adjustment, low de...
详细信息
ISBN:
(纸本)9781510829039
Low Density Parity Check(LDPC) code itself has a good performance and the application of LDPC in the communication field is extensive. LDPC codes have the characteristics of low bit error rate, easy adjustment, low decoding complexity and excellent decoding performance. LDPC is considered to be the "best performance" in the field of coding and decoding[1]. In this paper, we mainly introduce the theory of LDPC codes. And the use of the method of the LDPC code to do a further explanation. Several simple decoding schemes of LDPC are studied. Finally, using the method of function approximation to replace the confidence function to improve the bp LLR algorithm, and simulation experiments are carried out.
For solving a class of non-smooth unconstrained global optimization problems, we present a novel definition of the modified tunneling function which combines the characters of tunneling function and filled function, a...
详细信息
For solving a class of non-smooth unconstrained global optimization problems, we present a novel definition of the modified tunneling function which combines the characters of tunneling function and filled function, and then give a one-parameter modified tunneling function. Issues covered in the presented work include: theoretical properties, solution algorithms and numerical experiments. Furthermore, an improved artificial neural network hydrological forecasting method using the modified tunneling function is also reported. The preliminary experiment results confirm that the proposed approach is promising. (C) 2015 Elsevier Inc. All rights reserved.
As a class of important classifiers, feedforward neural networks (FNNs) have been used considerably in the study of pattern recognition. Since the inputs to FNNs are usually vectors, and many data are usually presente...
详细信息
As a class of important classifiers, feedforward neural networks (FNNs) have been used considerably in the study of pattern recognition. Since the inputs to FNNs are usually vectors, and many data are usually presented in the form of matrices, the matrices have to be decomposed into vectors before FNNs are employed. A drawback to this approach is that important information regarding correlations of elements within the original matrices are lost. Unlike traditional vector input based FNNs, a new algorithm of extended FNN with matrix inputs, called two-dimensional back-propagation (2D-bp), is proposed in this paper to classify matrix data directly, which utilizes the technique of incremental gradient descent to fully train the extended FNNs. These kinds of FNNs help to maintain the matrix structure of the 2D input features, which helps with image recognition. Promising experimental results of handwritten digits and face-image classification are provided to demonstrate the effectiveness of the proposed method. (C) 2015 Elsevier Ltd. All rights reserved.
暂无评论