The Experts below are selected from a list of 321 Experts worldwide ranked by ideXlab platform

Sheng Chen - One of the best experts on this subject based on the ideXlab platform.

  • CDC - A tunable radial basis function model for nonlinear system identification using particle swarm optimisation
    Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference, 2009
    Co-Authors: Sheng Chen, Xia Hong, B.l. Luk, Chris J. Harris
    Abstract:

    A tunable radial basis function (RBF) network model is proposed for nonlinear system identification using particle swarm optimisation (PSO). At each stage of orthogonal forward regression (OFR) model construction, PSO optimises one RBF unit's centre vector and Diagonal Covariance Matrix by minimising the leave-one-out (LOO) mean square error (MSE). This PSO aided OFR automatically determines how many tunable RBF nodes are sufficient for modelling. Compared with the-state-of-the-art local regularisation assisted orthogonal least squares algorithm based on the LOO MSE criterion for constructing fixed-node RBF network models, the PSO tuned RBF model construction produces more parsimonious RBF models with better generalisation performance and is computationally more efficient.

  • Construction of Tunable Radial Basis Function Networks Using Orthogonal Forward Selection
    IEEE transactions on systems man and cybernetics. Part B Cybernetics : a publication of the IEEE Systems Man and Cybernetics Society, 2008
    Co-Authors: Sheng Chen, Xia Hong, B.l. Luk, Chris J. Harris
    Abstract:

    An orthogonal forward selection (OFS) algorithm based on leave-one-out (LOO) criteria is proposed for the construction of radial basis function (RBF) networks with tunable nodes. Each stage of the construction process determines an RBF node, namely, its center vector and Diagonal Covariance Matrix, by minimizing the LOO statistics. For regression application, the LOO criterion is chosen to be the LOO mean-square error, while the LOO misclassification rate is adopted in two-class classification application. This OFS-LOO algorithm is computationally efficient, and it is capable of constructing parsimonious RBF networks that generalize well. Moreover, the proposed algorithm is fully automatic, and the user does not need to specify a termination criterion for the construction process. The effectiveness of the proposed RBF network construction procedure is demonstrated using examples taken from both regression and classification applications.

  • NARX-Based Nonlinear System Identification Using Orthogonal Least Squares Basis Hunting
    IEEE Transactions on Control Systems Technology, 2008
    Co-Authors: Sheng Chen, Xunxian Wang, Chris J. Harris
    Abstract:

    An orthogonal least squares technique for basis hunting (OLS-BH) is proposed to construct sparse radial basis function (RBF) models for NARX-type nonlinear systems. Unlike most of the existing RBF or kernel modelling methods, which places the RBF or kernel centers at the training input data points and use a fixed common variance for all the regressors, the proposed OLS-BH technique tunes the RBF center and Diagonal Covariance Matrix of individual regressor by minimizing the training mean square error. An efficient optimization method is adopted for this basis hunting to select regressors in an orthogonal forward selection procedure. Experimental results obtained using this OLS-BH technique demonstrate that it offers a state-of-the-art method for constructing parsimonious RBF models with excellent generalization performance.

  • Sparse support vector regression based on orthogonal forward selection for the generalised kernel model
    Neurocomputing, 2006
    Co-Authors: Xunxian Wang, David Lowe, Sheng Chen, Chris J. Harris
    Abstract:

    Abstract This paper considers sparse regression modelling using a generalised kernel model in which each kernel regressor has its individually tuned centre vector and Diagonal Covariance Matrix. An orthogonal least squares forward selection procedure is employed to select the regressors one by one, so as to determine the model structure. After the regressor selection, the corresponding model weight parameters are calculated from the Lagrange dual problem of the original regression problem with the regularised e -insensitive loss function. Unlike the support vector regression, this stage of the procedure involves neither reproducing kernel Hilbert space nor Mercer decomposition concepts. As the regressors used are not restricted to be positioned at training input points and each regressor has its own Diagonal Covariance Matrix, sparser representation can be obtained. Experiments involving one simulated example and three real data sets are used to demonstrate the effectiveness of the proposed novel regression modelling approach.

  • Parsimonious least squares support vector regression using orthogonal forward selection with the generalised kernel model
    International Journal of Modelling Identification and Control, 2006
    Co-Authors: Xunxian Wang, David Lowe, Sheng Chen, Chris J. Harris
    Abstract:

    A sparse regression modelling technique is developed using a generalised kernel model in which each kernel regressor has its individually tuned position (centre) vector and Diagonal Covariance Matrix. An orthogonal least squares forward selection procedure is employed to append the regressors one by one. After the determination of the model structure, namely the selection of an appropriate number of regressors, the model weight parameters are calculated from the Lagrange dual problem of the original least squares problem. Different from the least squares support vector regression, this regression modelling procedure involves neither reproducing kernel Hilbert space nor Mercer decomposition concepts. As the regressors used are not restricted to be positioned at training input points and each regressor has its own Diagonal Covariance Matrix, a very sparse representation can be obtained with excellent generalisation capability. Experimental results involving two real data sets demonstrate the effectiveness of the proposed regression modelling approach.

Chris J. Harris - One of the best experts on this subject based on the ideXlab platform.

  • CDC - A tunable radial basis function model for nonlinear system identification using particle swarm optimisation
    Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference, 2009
    Co-Authors: Sheng Chen, Xia Hong, B.l. Luk, Chris J. Harris
    Abstract:

    A tunable radial basis function (RBF) network model is proposed for nonlinear system identification using particle swarm optimisation (PSO). At each stage of orthogonal forward regression (OFR) model construction, PSO optimises one RBF unit's centre vector and Diagonal Covariance Matrix by minimising the leave-one-out (LOO) mean square error (MSE). This PSO aided OFR automatically determines how many tunable RBF nodes are sufficient for modelling. Compared with the-state-of-the-art local regularisation assisted orthogonal least squares algorithm based on the LOO MSE criterion for constructing fixed-node RBF network models, the PSO tuned RBF model construction produces more parsimonious RBF models with better generalisation performance and is computationally more efficient.

  • Construction of Tunable Radial Basis Function Networks Using Orthogonal Forward Selection
    IEEE transactions on systems man and cybernetics. Part B Cybernetics : a publication of the IEEE Systems Man and Cybernetics Society, 2008
    Co-Authors: Sheng Chen, Xia Hong, B.l. Luk, Chris J. Harris
    Abstract:

    An orthogonal forward selection (OFS) algorithm based on leave-one-out (LOO) criteria is proposed for the construction of radial basis function (RBF) networks with tunable nodes. Each stage of the construction process determines an RBF node, namely, its center vector and Diagonal Covariance Matrix, by minimizing the LOO statistics. For regression application, the LOO criterion is chosen to be the LOO mean-square error, while the LOO misclassification rate is adopted in two-class classification application. This OFS-LOO algorithm is computationally efficient, and it is capable of constructing parsimonious RBF networks that generalize well. Moreover, the proposed algorithm is fully automatic, and the user does not need to specify a termination criterion for the construction process. The effectiveness of the proposed RBF network construction procedure is demonstrated using examples taken from both regression and classification applications.

  • NARX-Based Nonlinear System Identification Using Orthogonal Least Squares Basis Hunting
    IEEE Transactions on Control Systems Technology, 2008
    Co-Authors: Sheng Chen, Xunxian Wang, Chris J. Harris
    Abstract:

    An orthogonal least squares technique for basis hunting (OLS-BH) is proposed to construct sparse radial basis function (RBF) models for NARX-type nonlinear systems. Unlike most of the existing RBF or kernel modelling methods, which places the RBF or kernel centers at the training input data points and use a fixed common variance for all the regressors, the proposed OLS-BH technique tunes the RBF center and Diagonal Covariance Matrix of individual regressor by minimizing the training mean square error. An efficient optimization method is adopted for this basis hunting to select regressors in an orthogonal forward selection procedure. Experimental results obtained using this OLS-BH technique demonstrate that it offers a state-of-the-art method for constructing parsimonious RBF models with excellent generalization performance.

  • Sparse support vector regression based on orthogonal forward selection for the generalised kernel model
    Neurocomputing, 2006
    Co-Authors: Xunxian Wang, David Lowe, Sheng Chen, Chris J. Harris
    Abstract:

    Abstract This paper considers sparse regression modelling using a generalised kernel model in which each kernel regressor has its individually tuned centre vector and Diagonal Covariance Matrix. An orthogonal least squares forward selection procedure is employed to select the regressors one by one, so as to determine the model structure. After the regressor selection, the corresponding model weight parameters are calculated from the Lagrange dual problem of the original regression problem with the regularised e -insensitive loss function. Unlike the support vector regression, this stage of the procedure involves neither reproducing kernel Hilbert space nor Mercer decomposition concepts. As the regressors used are not restricted to be positioned at training input points and each regressor has its own Diagonal Covariance Matrix, sparser representation can be obtained. Experiments involving one simulated example and three real data sets are used to demonstrate the effectiveness of the proposed novel regression modelling approach.

  • Parsimonious least squares support vector regression using orthogonal forward selection with the generalised kernel model
    International Journal of Modelling Identification and Control, 2006
    Co-Authors: Xunxian Wang, David Lowe, Sheng Chen, Chris J. Harris
    Abstract:

    A sparse regression modelling technique is developed using a generalised kernel model in which each kernel regressor has its individually tuned position (centre) vector and Diagonal Covariance Matrix. An orthogonal least squares forward selection procedure is employed to append the regressors one by one. After the determination of the model structure, namely the selection of an appropriate number of regressors, the model weight parameters are calculated from the Lagrange dual problem of the original least squares problem. Different from the least squares support vector regression, this regression modelling procedure involves neither reproducing kernel Hilbert space nor Mercer decomposition concepts. As the regressors used are not restricted to be positioned at training input points and each regressor has its own Diagonal Covariance Matrix, a very sparse representation can be obtained with excellent generalisation capability. Experimental results involving two real data sets demonstrate the effectiveness of the proposed regression modelling approach.

Mark J. F. Gales - One of the best experts on this subject based on the ideXlab platform.

  • NIPS - Factored Semi-Tied Covariance Matrices
    2000
    Co-Authors: Mark J. F. Gales
    Abstract:

    A new form of Covariance modelling for Gaussian mixture models and hidden Markov models is presented. This is an extension to an efficient form of Covariance modelling used in speech recognition, semi-tied Covariance matrices. In the standard form of semi-tied Covariance matrices the Covariance Matrix is decomposed into a highly shared decorrelating transform and a component-specific Diagonal Covariance Matrix. The use of a factored decorrelating transform is presented in this paper. This factoring effectively increases the number of possible transforms without increasing the number of free parameters. Maximum likelihood estimation schemes for all the model parameters are presented including the component/ transform assignment, transform and component parameters. This new model form is evaluated on a large vocabulary speech recognition task. It is shown that using this factored form of Covariance modelling reduces the word error rate.

  • Semi-tied Covariance matrices for hidden Markov models
    IEEE Transactions on Speech and Audio Processing, 1999
    Co-Authors: Mark J. F. Gales
    Abstract:

    There is normally a simple choice made in the form of the Covariance Matrix to be used with continuous-density HMMs. Either a Diagonal Covariance Matrix is used, with the underlying assumption that elements of the feature vector are independent, or a full or block-Diagonal Matrix is used, where all or some of the correlations are explicitly modeled. Unfortunately when using full or block-Diagonal Covariance matrices there tends to be a dramatic increase in the number of parameters per Gaussian component, limiting the number of components which may be robustly estimated. This paper introduces a new form of Covariance Matrix which allows a few "full" Covariance matrices to be shared over many distributions, whilst each distribution maintains its own "Diagonal" Covariance Matrix. In contrast to other schemes which have hypothesized a similar form, this technique fits within the standard maximum-likelihood criterion used for training HMMs. The new form of Covariance Matrix is evaluated on a large-vocabulary speech-recognition task. In initial experiments the performance of the standard system was achieved using approximately half the number of parameters. Moreover, a 10% reduction in word error rate compared to a standard system can be achieved with less than a 1% increase in the number of parameters and little increase in recognition time.

Ki Yong Lee - One of the best experts on this subject based on the ideXlab platform.

  • Local fuzzy PCA based GMM with dimension reduction on speaker identification
    Pattern Recognition Letters, 2004
    Co-Authors: Ki Yong Lee
    Abstract:

    To reduce the high dimensionality required for training of feature vectors in speaker identification, we propose an efficient GMM based on local PCA with fuzzy clustering. The proposed method firstly partitions the data space into several disjoint clusters by fuzzy clustering, and then performs PCA using the fuzzy Covariance Matrix on each cluster. Finally, the GMM for speaker is obtained from the transformed feature vectors with reduced dimension in each cluster. Compared to the conventional GMM with Diagonal Covariance Matrix, the proposed method shows faster result with less storage maintaining same performance.

  • ICCSA (2) - Efficient speaker identification based on robust VQ-PCA
    Computational Science and Its Applications — ICCSA 2003, 2003
    Co-Authors: Younjeong Lee, Joohun Lee, Ki Yong Lee
    Abstract:

    In this paper, and efficient speaker identification based on robust vector quantization principal component analysis (VQ-PCA) is proposed to solve the problems from outliers and high dimensionelity of training feature vectors in speaker identification. Firstly, the proposed method partitions the data space into several disjoint regions by roust VQ based on M-estimation. Secondly, the robust PCA is obtained from the Covariance Matrix in each region. Finally, our method obtains the Gaussian Mixture model (GMM) for speaker from the transformed feature vectors with reduced dimension by the robust PCA in each region. Compared to the conventional GMM with Diagonal Covariance Matrix, under the same performance, the proposed method gives faster results with less storage and, moreover, shows robust performance to outliers.

  • IDEAL - GMM Based on Local Fuzzy PCA for Speaker Identification
    Intelligent Data Engineering and Automated Learning, 2003
    Co-Authors: Jongjoo Lee, Jaeyeol Rheem, Ki Yong Lee
    Abstract:

    To reduce the high dimensionality required for training of feature vectors in speaker identification, we propose an efficient GMM based on local PCA with Fuzzy clustering. The proposed method firstly partitions the data space into several disjoint clusters by fuzzy clustering, and then performs PCA using the fuzzy Covariance Matrix in each cluster. Finally, the GMM for speaker is obtained from the transformed feature vectors with reduced dimension in each cluster. Compared to the conventional GMM with Diagonal Covariance Matrix, the proposed method needs less storage and shows faster result, under the same performance.

  • GMM based on local Fuzzy PCA for speaker identification
    Lecture Notes in Computer Science, 2003
    Co-Authors: Jongjoo Lee, Jaeyeol Rheem, Ki Yong Lee
    Abstract:

    To reduce the high dimensionality required for training of feature vectors in speaker identification, we propose an efficient GMM based on local PCA with Fuzzy clustering. The proposed method firstly partitions the data space into several disjoint clusters by fuzzy clustering, and then performs PCA using the fuzzy Covariance Matrix in each cluster. Finally, the GMM for speaker is obtained from the transformed feature vectors with reduced dimension in each cluster. Compared to the conventional GMM with Diagonal Covariance Matrix, the proposed method needs less storage and shows faster result, under the same performance.

  • GMM based on local PCA for speaker identification
    Electronics Letters, 2001
    Co-Authors: Changwoo Seo, Ki Yong Lee, Joohun Lee
    Abstract:

    An efficient Gaussian mixture modelling (GMM) method based on local principal component analysis (PCA) with vector quantisation (VQ) for speaker identification is proposed. The proposed method firstly partitions the data space into several disjoint regions by VQ, and then performs PCA in each region. Finally, the GMM for the speaker is obtained from the transformed feature vectors in each region. Compared to the conventional GMM method with Diagonal Covariance Matrix, under the same performance, the proposed method requires less storage and shows faster results.

Xunxian Wang - One of the best experts on this subject based on the ideXlab platform.

  • NARX-Based Nonlinear System Identification Using Orthogonal Least Squares Basis Hunting
    IEEE Transactions on Control Systems Technology, 2008
    Co-Authors: Sheng Chen, Xunxian Wang, Chris J. Harris
    Abstract:

    An orthogonal least squares technique for basis hunting (OLS-BH) is proposed to construct sparse radial basis function (RBF) models for NARX-type nonlinear systems. Unlike most of the existing RBF or kernel modelling methods, which places the RBF or kernel centers at the training input data points and use a fixed common variance for all the regressors, the proposed OLS-BH technique tunes the RBF center and Diagonal Covariance Matrix of individual regressor by minimizing the training mean square error. An efficient optimization method is adopted for this basis hunting to select regressors in an orthogonal forward selection procedure. Experimental results obtained using this OLS-BH technique demonstrate that it offers a state-of-the-art method for constructing parsimonious RBF models with excellent generalization performance.

  • Sparse support vector regression based on orthogonal forward selection for the generalised kernel model
    Neurocomputing, 2006
    Co-Authors: Xunxian Wang, David Lowe, Sheng Chen, Chris J. Harris
    Abstract:

    Abstract This paper considers sparse regression modelling using a generalised kernel model in which each kernel regressor has its individually tuned centre vector and Diagonal Covariance Matrix. An orthogonal least squares forward selection procedure is employed to select the regressors one by one, so as to determine the model structure. After the regressor selection, the corresponding model weight parameters are calculated from the Lagrange dual problem of the original regression problem with the regularised e -insensitive loss function. Unlike the support vector regression, this stage of the procedure involves neither reproducing kernel Hilbert space nor Mercer decomposition concepts. As the regressors used are not restricted to be positioned at training input points and each regressor has its own Diagonal Covariance Matrix, sparser representation can be obtained. Experiments involving one simulated example and three real data sets are used to demonstrate the effectiveness of the proposed novel regression modelling approach.

  • Parsimonious least squares support vector regression using orthogonal forward selection with the generalised kernel model
    International Journal of Modelling Identification and Control, 2006
    Co-Authors: Xunxian Wang, David Lowe, Sheng Chen, Chris J. Harris
    Abstract:

    A sparse regression modelling technique is developed using a generalised kernel model in which each kernel regressor has its individually tuned position (centre) vector and Diagonal Covariance Matrix. An orthogonal least squares forward selection procedure is employed to append the regressors one by one. After the determination of the model structure, namely the selection of an appropriate number of regressors, the model weight parameters are calculated from the Lagrange dual problem of the original least squares problem. Different from the least squares support vector regression, this regression modelling procedure involves neither reproducing kernel Hilbert space nor Mercer decomposition concepts. As the regressors used are not restricted to be positioned at training input points and each regressor has its own Diagonal Covariance Matrix, a very sparse representation can be obtained with excellent generalisation capability. Experimental results involving two real data sets demonstrate the effectiveness of the proposed regression modelling approach.

  • Orthogonal least square with boosting for regression
    Lecture Notes in Computer Science, 2004
    Co-Authors: Sheng Chen, Xunxian Wang, David Brown
    Abstract:

    A novel technique is presented to construct sparse regression models based on the orthogonal least square method with boosting. This technique tunes the mean vector and Diagonal Covariance Matrix of individual regressor by incrementally minimizing the training mean square error. A weighted optimization method is developed based on boosting to append regressors one by one in an orthogonal forward selection procedure. Experimental results obtained using this technique demonstrate that it offers a viable alternative to the existing state-of-art kernel modeling methods for constructing parsimonious regression models.

  • IDEAL - Orthogonal Least Square with Boosting for Regression
    Lecture Notes in Computer Science, 2004
    Co-Authors: Sheng Chen, Xunxian Wang, David Brown
    Abstract:

    A novel technique is presented to construct sparse regression models based on the orthogonal least square method with boosting. This technique tunes the mean vector and Diagonal Covariance Matrix of individual regressor by incrementally minimizing the training mean square error. A weighted optimization method is developed based on boosting to append regressors one by one in an orthogonal forward selection procedure. Experimental results obtained using this technique demonstrate that it offers a viable alternative to the existing state-of-art kernel modeling methods for constructing parsimonious regression models.