Tensor representation is useful to reduce the overfitting problem in vector-based learning algorithm in pattern recognition.This is mainly because the structure information of objects in pattern analysis is a reasonab...Tensor representation is useful to reduce the overfitting problem in vector-based learning algorithm in pattern recognition.This is mainly because the structure information of objects in pattern analysis is a reasonable constraint to reduce the number of unknown parameters used to model a classifier.In this paper, we generalize the vector-based learning algorithm TWin Support Vector Machine(TWSVM) to the tensor-based method TWin Support Tensor Machines(TWSTM), which accepts general tensors as input.To examine the effectiveness of TWSTM, we implement the TWSTM method for Microcalcification Clusters(MCs) detection.In the tensor subspace domain, the MCs detection procedure is formulated as a supervised learning and classification problem, and TWSTM is used as a classifier to make decision for the presence of MCs or not.A large number of experiments were carried out to evaluate and compare the performance of the proposed MCs detection algorithm.By comparison with TWSVM, the tensor version reduces the overfitting problem.展开更多
This study quantifies the main characteristics of a terrain-following, G-coordinate through mathematical analyses of its covariant and contravariant basis vectors as well as the vertical coordinate of σ. A 3-D schema...This study quantifies the main characteristics of a terrain-following, G-coordinate through mathematical analyses of its covariant and contravariant basis vectors as well as the vertical coordinate of σ. A 3-D schematic of the σ-coordinate in a curvilinear coordinate system is provided in this study. The characteristics of the basis vectors were broken down into their ＂local vector charac- teristics＂ and ＂spatial distribution characteristics＂, and the exact expressions of the covariant; in addition, the con- travariant basis vectors of the G-coordinate used to eluci- date their detailed characteristics were properly solved. Through rewriting the expression of the vertical coordi- nate of G, a mathematical expression of all the cr-coor- dinate surfaces was found, thereby quantifying the so- called terrain-following characteristics and lack of flexi- bility to adjust the slope variation of G-coordinate sur- faces for the classic definition of G. Finally, an analysis on the range value of the vertical coordinate demonstrated that the general value range of G could be obtained by eliminating the G-coordinate surfaces below the Earth＇s surface. All these quantitative descriptions of the charac- teristics of G-coordinate were the foundation for improv- ing the G-coordinate or creating a new one.展开更多
基金Supported by the National Natural Science Foundation of China (No. 60771068)the Natural Science Basic Research Plan in Shaanxi Province of China (No. 2007F248)
文摘Tensor representation is useful to reduce the overfitting problem in vector-based learning algorithm in pattern recognition.This is mainly because the structure information of objects in pattern analysis is a reasonable constraint to reduce the number of unknown parameters used to model a classifier.In this paper, we generalize the vector-based learning algorithm TWin Support Vector Machine(TWSVM) to the tensor-based method TWin Support Tensor Machines(TWSTM), which accepts general tensors as input.To examine the effectiveness of TWSTM, we implement the TWSTM method for Microcalcification Clusters(MCs) detection.In the tensor subspace domain, the MCs detection procedure is formulated as a supervised learning and classification problem, and TWSTM is used as a classifier to make decision for the presence of MCs or not.A large number of experiments were carried out to evaluate and compare the performance of the proposed MCs detection algorithm.By comparison with TWSVM, the tensor version reduces the overfitting problem.
基金supported by the National Natural Science Foundation of China under Grant Nos. 40821092,40633016,and 40875022
文摘This study quantifies the main characteristics of a terrain-following, G-coordinate through mathematical analyses of its covariant and contravariant basis vectors as well as the vertical coordinate of σ. A 3-D schematic of the σ-coordinate in a curvilinear coordinate system is provided in this study. The characteristics of the basis vectors were broken down into their ＂local vector charac- teristics＂ and ＂spatial distribution characteristics＂, and the exact expressions of the covariant; in addition, the con- travariant basis vectors of the G-coordinate used to eluci- date their detailed characteristics were properly solved. Through rewriting the expression of the vertical coordi- nate of G, a mathematical expression of all the cr-coor- dinate surfaces was found, thereby quantifying the so- called terrain-following characteristics and lack of flexi- bility to adjust the slope variation of G-coordinate sur- faces for the classic definition of G. Finally, an analysis on the range value of the vertical coordinate demonstrated that the general value range of G could be obtained by eliminating the G-coordinate surfaces below the Earth＇s surface. All these quantitative descriptions of the charac- teristics of G-coordinate were the foundation for improv- ing the G-coordinate or creating a new one.
