We improve the twin support vector machine(TWSVM)to be a novel nonparallel hyperplanes classifier,termed as ITSVM(improved twin support vector machine),for binary classification.By introducing the diferent Lagrangian ...We improve the twin support vector machine(TWSVM)to be a novel nonparallel hyperplanes classifier,termed as ITSVM(improved twin support vector machine),for binary classification.By introducing the diferent Lagrangian functions for the primal problems in the TWSVM,we get an improved dual formulation of TWSVM,then the resulted ITSVM algorithm overcomes the common drawbacks in the TWSVMs and inherits the essence of the standard SVMs.Firstly,ITSVM does not need to compute the large inverse matrices before training which is inevitable for the TWSVMs.Secondly,diferent from the TWSVMs,kernel trick can be applied directly to ITSVM for the nonlinear case,therefore nonlinear ITSVM is superior to nonlinear TWSVM theoretically.Thirdly,ITSVM can be solved efciently by the successive overrelaxation(SOR)technique or sequential minimization optimization(SMO)method,which makes it more suitable for large scale problems.We also prove that the standard SVM is the special case of ITSVM.Experimental results show the efciency of our method in both computation time and classification accuracy.展开更多
Minimax programming problems involving generalized （p, r）-invex functions are consid- ered. Parametric sufficient optimality conditions and duality results are established under the aforesaid assumptions on the obje...Minimax programming problems involving generalized （p, r）-invex functions are consid- ered. Parametric sufficient optimality conditions and duality results are established under the aforesaid assumptions on the objective and constraint functions.展开更多
For multi-cell curve box girder, the finite strip governing equation was derived on the basis of Novozhilov theory and orthogonal property of harmonious function series. Dynamic Bayesian error function of mechanical p...For multi-cell curve box girder, the finite strip governing equation was derived on the basis of Novozhilov theory and orthogonal property of harmonious function series. Dynamic Bayesian error function of mechanical parameters of multi-cell curve box girder was achieved with Bayesian statistical theory. The corresponding formulas of dynamic Bayesian expectation and variance were obtained. After the one-dimensional optimization search method for the automatic determination of step length of the mechanical parameter was put forward, the optimization identification calculative formulas were also obtained by adopting conjugate gradient method. Then the steps of dynamic Bayesian identification of mechanical parameters of multi-cell curve box girder were stated in detail. Through analysis of a classic example, the dynamic Bayesian identification processes of mechanical parameters are steadily convergent to the true values, which proves that dynamic Bayesian theory and conjugate gradient theory are suitable for the identification calculation and the compiled procedure is correct. It is of significance that the foreknown information of mechanical parameters should be set with reliable practical engineering experiences instead of arbitrary selection.展开更多
基金supported by National Natural Science Foundation of China(Grant Nos.11271361 and 70921061)the CAS/SAFEA International Partnership Program for Creative Research Teams,Major International(Regional)Joint Research Project(Grant No.71110107026)+1 种基金the Ministry of Water Resources Special Funds for Scientific Research on Public Causes(Grant No.201301094)Hong Kong Polytechnic University(Grant No.B-Q10D)
文摘We improve the twin support vector machine(TWSVM)to be a novel nonparallel hyperplanes classifier,termed as ITSVM(improved twin support vector machine),for binary classification.By introducing the diferent Lagrangian functions for the primal problems in the TWSVM,we get an improved dual formulation of TWSVM,then the resulted ITSVM algorithm overcomes the common drawbacks in the TWSVMs and inherits the essence of the standard SVMs.Firstly,ITSVM does not need to compute the large inverse matrices before training which is inevitable for the TWSVMs.Secondly,diferent from the TWSVMs,kernel trick can be applied directly to ITSVM for the nonlinear case,therefore nonlinear ITSVM is superior to nonlinear TWSVM theoretically.Thirdly,ITSVM can be solved efciently by the successive overrelaxation(SOR)technique or sequential minimization optimization(SMO)method,which makes it more suitable for large scale problems.We also prove that the standard SVM is the special case of ITSVM.Experimental results show the efciency of our method in both computation time and classification accuracy.
文摘Minimax programming problems involving generalized （p, r）-invex functions are consid- ered. Parametric sufficient optimality conditions and duality results are established under the aforesaid assumptions on the objective and constraint functions.
基金supported by the National Natural Science Foundation of China (Grant Nos. 10772078 and 11072108)the Transportation Science Foundation of Jiangsu Province (Grant No. 09Y012)
文摘For multi-cell curve box girder, the finite strip governing equation was derived on the basis of Novozhilov theory and orthogonal property of harmonious function series. Dynamic Bayesian error function of mechanical parameters of multi-cell curve box girder was achieved with Bayesian statistical theory. The corresponding formulas of dynamic Bayesian expectation and variance were obtained. After the one-dimensional optimization search method for the automatic determination of step length of the mechanical parameter was put forward, the optimization identification calculative formulas were also obtained by adopting conjugate gradient method. Then the steps of dynamic Bayesian identification of mechanical parameters of multi-cell curve box girder were stated in detail. Through analysis of a classic example, the dynamic Bayesian identification processes of mechanical parameters are steadily convergent to the true values, which proves that dynamic Bayesian theory and conjugate gradient theory are suitable for the identification calculation and the compiled procedure is correct. It is of significance that the foreknown information of mechanical parameters should be set with reliable practical engineering experiences instead of arbitrary selection.
