Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 39 for “"Gram matrix"”.

  1. Efficient and Accurate Machine Learning Model Computation in Data Science Languages with Data Summarization via a Gram Matrix

    Nowadays, data science analysts prefer “easy” high-level languages for machine learning computation like R and Python, but they present memory and speed limitations. Also, scalability is another issue when the data set size grows. Data summarization has been a fundamental technique in data mining …

    houston Repository record for Efficient and Accurate Machine Learning Model Computation in Data Science Languages with Data Summarization via a Gram Matrix (opens in a new tab)

  2. Hyperbolic Geometry and Reflection Groups

    … such sequence we associate a square symmetric matrix, the Gram matrix, which gives information about angle and incidence properties of the hyperplanes. We find when a given matrix is the Gram matrix of some sequence of hyperplanes, and when a sequence is determined up to isometry by its Gram …

    auckland-ms Repository record for Hyperbolic Geometry and Reflection Groups (opens in a new tab)

  3. Detection algorithms for spatial data

    … directly as a function of the centered kernel Gram matrix. Subsequently, a methodology for the fast computation of the centered kernel Gram matrix is proposed. The key idea behind the proposed methodology is to decompose the set of image pixels into clusters, and expediting the computations by …

    must-thes Repository record for Detection algorithms for spatial data (opens in a new tab)

  4. PATTERN RECOGNITION IN CLASS IMBALANCED DATASETS

    … modifying kernel function to generate the gram matrix and optimizing the NN structure. First, the training dataset is extracted from the available sample set through an iterative process of selective under-sampling. Then, the proposed artificial NN comprises of a kernel function optimizer …

    vcu Repository record for PATTERN RECOGNITION IN CLASS IMBALANCED DATASETS (opens in a new tab)

  5. Multiplicative distance functions

    … (s) can be represented as the determinant of a Gram matrix in a Hilbert space dependent on s and Φ. This revelation allows us to write HN (s) as the product of the norms of N vectors in the associated Hilbert space. Several examples are presented. Similarly, when N is even we introduce a …

    texas Repository record for Multiplicative distance functions (opens in a new tab)

  6. Low Complexity Detection and Precoding for Massive MIMO Systems: Algorithm, Architecture, and Application

    … VLSI architecture to perform both the necessary Gram matrix computation and Neumann series based matrix inversion. The Neumann series approach, however, suffers from a considerable error-rate performance loss if the ratio of BS to user antennas is not large enough. To improve the performance, we …

    rice Repository record for Low Complexity Detection and Precoding for Massive MIMO Systems: Algorithm, Architecture, and Application (opens in a new tab)

  7. Mobius Structures, Einstein Metrics, and Discrete Conformal Variations on Piecewise Flat Two and Three Dimensional Manifolds

    … faces. These characterizations analyze the Gram matrix, a matrix with entries given by cosines of dihedral angles. Hyperideal hyperbolic simplices are non-compact generalizations of hyperbolic simplices wherein the vertices lie outside hyperbolic space. We extend recent characterization …

    arizona-thes Repository record for Mobius Structures, Einstein Metrics, and Discrete Conformal Variations on Piecewise Flat Two and Three Dimensional Manifolds (opens in a new tab)

  8. Digital Filter Design Using Semidefinite Programming

    … can be represented by a positive semidefinite matrix known as Gram matrix representation. This allows us to utilize the framework of linear conic optimization, especially semidefinite programming to obtain filters based on given specifications and optimal with respect to some property of the …

    chalmers Repository record for Digital Filter Design Using Semidefinite Programming (opens in a new tab)

  9. Orthogonality and convergence of discrete Zernike polynomials

    … the orthonormal polynomials visually with the Gram matrix and computationally with the rank and condition number. The convergence of the Fourier-Zernike coe\ufb03cients and the Fourier-Zernike series are also examined using various measures of error. The orthogonality and convergence are …

    unm Repository record for Orthogonality and convergence of discrete Zernike polynomials (opens in a new tab)

  10. Jacobi-SVD Channel Estimation for Amplify and Forward MIMO Relay

    … by computing the Eigen decomposition of gram matrix have proven to be very difficult and impractical to implement due to associated numerical difficulties. Therefore Jacobi Singular Value Decomposition parallel algorithm for channel decomposition is proposed. Performance evaluation is …

    calgary Repository record for Jacobi-SVD Channel Estimation for Amplify and Forward MIMO Relay (opens in a new tab)

  11. Statistical Inference and Computational Methods on Graphs with Geometric Structures

    … the identifiable target is a rectangular cross-Gram matrix. It gives rate--distortion lower bounds governed by the effective dimension of the rectangular low-rank variety and a spectral recovery guarantee based on the symmetric dilation of the centered biadjacency matrix. Together, these results …

    gatech Repository record for Statistical Inference and Computational Methods on Graphs with Geometric Structures (opens in a new tab)

  12. The Distribution of the Cusped Hypocycloidal Mahler Measure

    … Hd(s) may be represented as the determinant of a Gram matrix in a Hilbert space determined by Phi and s. We thus discover properties of Hd(mu^(N); s) as a rational function of s.

    oregon Repository record for The Distribution of the Cusped Hypocycloidal Mahler Measure (opens in a new tab)

  13. Convolutional neural network architectures and training improvements: visual emotion recognition case /

    … main visual representation learned by the CNN, Gram matrix modules are added to capture stylistic information from intermediate feature maps, such as texture and color-related patterns. In addition, contrastive-center loss is incorporated into training to improve the separation of emotion …

    vilnius Repository record for Convolutional neural network architectures and training improvements: visual emotion recognition case / (opens in a new tab)

  14. Analytic Solutions to the NLS, MKDV, and KP Equations

    … are obtained by taking the determinant of a Gram matrix of generalized Schur polynomials. A classification system of KP-1 solitons is implemented in terms of the degree of the generalized Schur polynomial, known as the depth, and the spectral parameter. Known results are recreated and two …

    arizona-thes Repository record for Analytic Solutions to the NLS, MKDV, and KP Equations (opens in a new tab)

  15. Some Representation Theory Of The Dilute Blob Algebra

    … of the Motzkin algebra. After we define basis diagrams of the dilute blob algebra, we give generators for the dilute blob algebra. A bijection between basis diagrams of the dilute blob algebra and basis diagrams of the left-right symmetric Motzkin algebra is also studied. We prove that the dilute …

    whiterose Download PDF for Some Representation Theory Of The Dilute Blob Algebra (opens in a new tab)

  16. The Distance to Uncontrollability via Linear Matrix Inequalities

    … two alternative algorithms that result in linear matrix inequalities. For the first algorithm, proposed by Ebihara et. al., a semidefinite programming problem is derived via the Kalman-Yakubovich-Popov (KYP) lemma. The dual formulation is also considered and leads to rank conditions for exactness …

    vt Repository record for The Distance to Uncontrollability via Linear Matrix Inequalities (opens in a new tab)

  17. ShuFFLE: Automated Framework for HArdware Accelerated Iterative Big Data Analysis

    … acceleration of large and dense (non-sparse) Gram matrices. Such matrices arise in most contemporary data mining; they are hard to handle because of the complexity of known matrix transformation algorithms and the inseparability of non-sparse correlations. ShuFFLE learns the properties of the …

    rice Repository record for ShuFFLE: Automated Framework for HArdware Accelerated Iterative Big Data Analysis (opens in a new tab)

  18. Frames with desired angle properties

    … the same number of times in each row of the Gram matrix. We call this type of set a regular two-distance set. We present various properties of such sets as well as focus on the case where the sets form tight frames for the underlying space. Several connections among regular two-distance sets, …

    missouri Repository record for Frames with desired angle properties (opens in a new tab)

  19. Adaptive Constraint Reduction for Convex Quadratic Programming and Training Support Vector Machines

    Convex quadratic programming (CQP) is an optimization problem of minimizing a convex quadratic objective function subject to linear constraints. We propose an adaptive constraint reduction primal-dual interior-point algorithm for convex quadratic programming with many more constraints than …

    maryland Repository record for Adaptive Constraint Reduction for Convex Quadratic Programming and Training Support Vector Machines (opens in a new tab)

  20. Non-asymptotic properties of spectral decomposition of large gram-type matrices with applications to high-dimensional inference

    … of spectral decomposition of high-dimensional Gram-type matrices based on factor models. Specifically, we derive the exponential tail bound for the first and second moments of the deviation between the empirical and population eigenvectors to the right Gram matrix as well as the Berry-Esseen …

    colostate Repository record for Non-asymptotic properties of spectral decomposition of large gram-type matrices with applications to high-dimensional inference (opens in a new tab)

Page 1 of 2