Search

There should be different requirements for False Reject rate and False Accept rate in classification applications, and classifier learning should use an asymmetric factor to balance between False Reject rate and False Accept rate. A novel AdaBoost algorithm was developed with the asymmetric weight. Moreover we provide the theoretical analysis of its performance and derive the upper bound of the classification error.

In this paper the problems of parameter estimation and order determination of an exponential (EX) model are studied in the time domain. In order to estimate the parameters, the parameter equations of an EX model are given in terms of the autocorrelation function, which is similar to the Yule-Walker equations of an autoregressive moving-average model. Estimates of parameters are obtained with the aid of the parameter equations and theorems are proved relating the convergence rate and asymptotic distribution of the estimates. We present two kinds of methods for estimating the order and prove that the estimates of the order are consistent.

Traditional manifold learning algorithms, such as ISOMAP, LLE, and Laplacian Eigenmap, mainly focus on uncovering the latent low-dimensional geometry structure of the training samples in an unsupervised manner where useful class information is ignored. Therefore, the derived low-dimensional representations are not necessarily optimal in discriminative capability. In this paper, we study the discriminant analysis problem by considering the nonlinear manifold structure of data space. To this end, firstly, a new clustering algorithm, called Intra-Cluster Balanced K-Means (ICBKM), is proposed to partition the samples into multiple clusters while ensure that there are balanced samples for the classes within each cluster; approximately, each cluster can be considered as a local patch on the embedded manifold. Then, the local discriminative projections for different clusters are simultaneously calculated by optimizing the global Fisher Criterion based on the cluster weighted data representation. Compared with traditional linear/kernel discriminant analysis (KDA) algorithms, our proposed algorithm has the following characteristics: 1) it essentially is a KDA algorithm with specific geometry-adaptive-kernel tailored to the specific data structure, in contrast to traditional KDA in which the kernel is fixed and independent to the data set; 2) it is approximately a locally linear while globally nonlinear discriminant analyzer; 3) it does not need to store the original samples for computing the low-dimensional representation of a new data; and 4) it is computationally efficient compared with traditional KDA when the sample number is large. The toy problem on artificial data demonstrates the effectiveness of our proposed algorithm in deriving discriminative representations for problems with nonlinear classification hyperplane. The face recognition experiments on YALE and CMU PIE databases show that our proposed algorithm significantly outperforms linear discriminant analysis (LDA) as well as Mixture LDA, and has higher accuracy than KDA with traditional kernels