The Experts below are selected from a list of 34842 Experts worldwide ranked by ideXlab platform
Junmo Kim - One of the best experts on this subject based on the ideXlab platform.
-
unsupervised simultaneous Orthogonal Basis clustering feature selection
Computer Vision and Pattern Recognition, 2015Co-Authors: Dongyoon Han, Junmo KimAbstract:In this paper, we propose a novel unsupervised feature selection method: Simultaneous Orthogonal Basis Clustering Feature Selection (SOCFS). To perform feature selection on unlabeled data effectively, a regularized regression-based formulation with a new type of target matrix is designed. The target matrix captures latent cluster centers of the projected data points by performing Orthogonal Basis clustering, and then guides the projection matrix to select discriminative features. Unlike the recent unsupervised feature selection methods, SOCFS does not explicitly use the pre-computed local structure information for data points represented as additional terms of their objective functions, but directly computes latent cluster information by the target matrix conducting Orthogonal Basis clustering in a single unified term of the proposed objective function. It turns out that the proposed objective function can be minimized by a simple optimization algorithm. Experimental results demonstrate the effectiveness of SOCFS achieving the state-of-the-art results with diverse real world datasets.
-
CVPR - Unsupervised Simultaneous Orthogonal Basis Clustering Feature Selection
2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015Co-Authors: Dongyoon Han, Junmo KimAbstract:In this paper, we propose a novel unsupervised feature selection method: Simultaneous Orthogonal Basis Clustering Feature Selection (SOCFS). To perform feature selection on unlabeled data effectively, a regularized regression-based formulation with a new type of target matrix is designed. The target matrix captures latent cluster centers of the projected data points by performing Orthogonal Basis clustering, and then guides the projection matrix to select discriminative features. Unlike the recent unsupervised feature selection methods, SOCFS does not explicitly use the pre-computed local structure information for data points represented as additional terms of their objective functions, but directly computes latent cluster information by the target matrix conducting Orthogonal Basis clustering in a single unified term of the proposed objective function. It turns out that the proposed objective function can be minimized by a simple optimization algorithm. Experimental results demonstrate the effectiveness of SOCFS achieving the state-of-the-art results with diverse real world datasets.
Guanhua Chen - One of the best experts on this subject based on the ideXlab platform.
-
time dependent density functional theory quantum transport simulation in non Orthogonal Basis
Journal of Chemical Physics, 2013Co-Authors: Yan Ho Kwok, Hang Xie, Chiyung Yam, Xiao Zheng, Guanhua ChenAbstract:Basing on the earlier works on the hierarchical equations of motion for quantum transport, we present in this paper a first principles scheme for time-dependent quantum transport by combining time-dependent density functional theory (TDDFT) and Keldysh's non-equilibrium Green's function formalism. This scheme is beyond the wide band limit approximation and is directly applicable to the case of non-Orthogonal Basis without the need of Basis transformation. The overlap between the Basis in the lead and the device region is treated properly by including it in the self-energy and it can be shown that this approach is equivalent to a lead-device Orthogonalization. This scheme has been implemented at both TDDFT and density functional tight-binding level. Simulation results are presented to demonstrate our method and comparison with wide band limit approximation is made. Finally, the sparsity of the matrices and computational complexity of this method are analyzed.
Saburo Kakei - One of the best experts on this subject based on the ideXlab platform.
-
An Orthogonal Basis for the -type Calogero model
Journal of Physics A: Mathematical and General, 1997Co-Authors: Saburo KakeiAbstract:We investigate algebraic structure for the -type Calogero model by using the exchange-operator formalism. We show that the set of Jack polynomials whose arguments are Dunkl-type operators provides an Orthogonal Basis.
-
An Orthogonal Basis for the $B_N$-type Calogero model
arXiv: Exactly Solvable and Integrable Systems, 1996Co-Authors: Saburo KakeiAbstract:We investigate algebraic structure for the $B_N$-type Calogero model by using the exchange-operator formalism. We show that the set of the Jack polynomials whose arguments are Dunkl-type operators provides an Orthogonal Basis.
Dongyoon Han - One of the best experts on this subject based on the ideXlab platform.
-
unsupervised simultaneous Orthogonal Basis clustering feature selection
Computer Vision and Pattern Recognition, 2015Co-Authors: Dongyoon Han, Junmo KimAbstract:In this paper, we propose a novel unsupervised feature selection method: Simultaneous Orthogonal Basis Clustering Feature Selection (SOCFS). To perform feature selection on unlabeled data effectively, a regularized regression-based formulation with a new type of target matrix is designed. The target matrix captures latent cluster centers of the projected data points by performing Orthogonal Basis clustering, and then guides the projection matrix to select discriminative features. Unlike the recent unsupervised feature selection methods, SOCFS does not explicitly use the pre-computed local structure information for data points represented as additional terms of their objective functions, but directly computes latent cluster information by the target matrix conducting Orthogonal Basis clustering in a single unified term of the proposed objective function. It turns out that the proposed objective function can be minimized by a simple optimization algorithm. Experimental results demonstrate the effectiveness of SOCFS achieving the state-of-the-art results with diverse real world datasets.
-
CVPR - Unsupervised Simultaneous Orthogonal Basis Clustering Feature Selection
2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015Co-Authors: Dongyoon Han, Junmo KimAbstract:In this paper, we propose a novel unsupervised feature selection method: Simultaneous Orthogonal Basis Clustering Feature Selection (SOCFS). To perform feature selection on unlabeled data effectively, a regularized regression-based formulation with a new type of target matrix is designed. The target matrix captures latent cluster centers of the projected data points by performing Orthogonal Basis clustering, and then guides the projection matrix to select discriminative features. Unlike the recent unsupervised feature selection methods, SOCFS does not explicitly use the pre-computed local structure information for data points represented as additional terms of their objective functions, but directly computes latent cluster information by the target matrix conducting Orthogonal Basis clustering in a single unified term of the proposed objective function. It turns out that the proposed objective function can be minimized by a simple optimization algorithm. Experimental results demonstrate the effectiveness of SOCFS achieving the state-of-the-art results with diverse real world datasets.
Ester Pérez Sinusía - One of the best experts on this subject based on the ideXlab platform.
-
Erratum on "Orthogonal Basis with a conicoid first mode for shape specification of optical surfaces".
Optics Express, 2016Co-Authors: Chelo Ferreira, José L López, Rafael Navarro, Ester Pérez SinusíaAbstract:We found an error affecting the results presented in Figure 7 of our article "Orthogonal Basis with a conicoid first mode for shape specification of optical surfaces". Here we publish the revised Fig. 7.
-
Orthogonal Basis with a conicoid first mode for shape specification of optical surfaces: reply.
Optics Express, 2016Co-Authors: Chelo Ferreira, José L López, Rafael Navarro, Ester Pérez SinusíaAbstract:We present some comments to the paper "Orthogonal Basis with a conicoid first mode for shape specification of optical surfaces: comment."
