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 26 for “"hierarchical matrices"”.
-
Eigenvalue Algorithms for Symmetric Hierarchical Matrices
… computation of eigenvalues of symmetric hierarchical matrices. The numerical algorithms used for this computation are derivations of the LR Cholesky algorithm, the preconditioned inverse iteration, and a bisection method based on LDLT factorizations. The investigation of QR decompositions …
-
The science of high performance algorithms for hierarchical matrices
Many matrices in scientific computing, statistical inference, and machine learning exhibit sparse and low-rank structure. Typically, such structure is exposed by appropriate matrix permutation of rows and columns, and exploited by constructing an hierarchical approximation. That is, the matrix can …
-
High-Performance Software for Quantum Chemistry and Hierarchical Matrices
… simulations. Notably, we use preconditioned hierarchical matrices for the hyperparameter optimization and prediction phases of Gaussian process regression, develop a sparse triple matrix product on GPUs, and investigate 3D matrix-matrix multiplications for Chebyshev-filtered subspace …
-
H-matrix preconditioning for the time-harmonic electromagnetic discontinuous Galerkin method
Hierarchical matrices, or H-matrices are an error-controllable framework that permit efficient inversion and decomposition of matrices arising in time-harmonic electromagnetics applications. This work evaluates the capabilities of H-matrices for preconditioning iterative solutions to the …
-
Fast direct algorithms for elliptic equations via hierarchical matrix compression
… (2009) on combining the multifrontal method with hierarchical matrices. We offer a more geometric interpretation of that approach, extend it in two dimensions to the unstructured mesh case, and detail an adaptive decomposition procedure for selectively refined meshes. Linear time complexity is …
-
Structured matrix recovery and approximation from matrix-vector products
… such as tridiagonal, Toeplitz, Hankel, low-rank, hierarchical, and sparse matrices, are provided. For sparse and hierarchical matrices, the more general problem of near-optimal approximation from matrix-vector products is analyzed. This research is motivated by the field of operator learning, …
-
Data-driven discovery of Green's functions
… operators, and exploiting the low-rank hierarchical structure of Green's functions using hierarchical matrices.</p> <p>Rational neural networks (NNs) are introduced and consist of neural networks with trainable rational activation functions. The highly compositional structure of these …
-
Micromagnetic simulations of three dimensional core-shell nanostructures
… FEM/BEM by using a data-sparse matrix type (hierarchical matrices) in order to extend the range of structures accessible by micromagnetic simulations.<br/>It is shown that this approximation leads only to negligible errors. The performed micromagnetic simulations include the finding of …
-
Solution of large scale elastic problems with the boundary element method: applications to infrasound and micro-seismicity generated by wind turbines
… static as well as harmonic problems based on the hierarchical matrix methodology combined with the algebraic low rank approximation technique, known as Adaptive Cross Approximation (ACA), is proposed. For the efficient algebraic compression of the utilized hierarchical matrices, different versions …
-
Hierarchical, Bottom-up Assembly of Multi-Species Living Materials through Bacterial Secretion and Display
… and modify biopolymers, which assemble into hierarchical matrices. Inspired by this process, the nascent field of Engineered Living Materials (ELMs) uses synthetic biology to build novel materials with tunable, non-natural functions. However, to date ELMs have been largely produced via a …
-
Ανάπτυξη υπολογιστικών τεχνικών με τη χρήση διατυπώσεων μεικτών πεπερασμένων στοιχείων στο σύστημα των εξισώσεων του Maxwell για προβλήματα ηλεκτρομαγνητικής διάδοσης
… and easily constructible approaches. The use of hierarchical matrices technique for constructing preconditioners for the E-B formulation was explored, demonstrating the potential integration of hierarchical matrix techniques into practical preconditioning processes. The dissertation highlights …
-
Solution of a Second Order Elliptic Partial Differential Equation with Varying Complex Coefficients: An Application for Computing Effective Complex Electrical Properties of Materials represented by 3D Images
… The solution is computed using the technique of Hierarchical Matrices in combination with a Linear Method and the Generalised Minimal Residual Method algorithm. A C code was written to implement the scheme. The code uses the NetCDF library to read the 3D image and the H-Lib(Pro) library to work …
-
Efficient numerical treamtent of the fractional Laplacian in three dimensions
… For the second problem, we choose to use hierarchical matrices (H -matrices) and prove that this method is also efficient for the fractional Laplacian. H -matrices offer the advantage that the effort for constructing the approximation and performing the matrix-vector-multiplication is only …
-
Randomized algorithms for the efficient solution of elliptic PDEs on modern architectures
… The unifying theme of the work is the use of hierarchical matrices (𝓗-matrices) to accelerate key compute kernels. The term 𝓗-matrix refers to a matrix that can hierarchically be tessellated into submatrices in such a way that each submatrix is either of low numerical rank, or small enough …
-
Über die Lösung von elliptischen Randwertproblemen mittels Gebietszerlegungstechniken, Hierarchischer Matrizen und der Methode der finiten Elemente
… durch den sogenannten HDD-Algorithmus (engl. hierarchical domain decomposition). Die meisten der zu erstellenden Matrizen sind dabei vollbesetzt, weshalb wir sie mittels Hierarchischer Matrizen approximieren. Mit Hilfe der Hierarchischen Matrizen können wir die Matrizen mit einem fast linearen …
-
Kernel methods: generalisations, scalability and towards the future of machine learning
… has been ample work on approximating kernel matrices via low-rank approximations, Bayesian committee machines and hierarchical matrices to name but a few approaches, little work has been investigated in the area of stochastic trace estimation. This work thoroughly explores this avenue. It …
-
On Multigrid and H-Matrix Methods for Partial Integro-Differential Equations
… this thesis is the development of multigrid and hierarchical matrix solution procedures with almost linear computational complexity for classes of partial integro-differential problems. An elliptic partial integro-differential equation, a convection-diffusion partial integro-differential equation …
-
Hierarchical Matrix Techniques on Massively Parallel Computers
Hierarchical matrix (H-matrix) techniques can be used to efficiently treat dense matrices. With an H-matrix, the storage requirements and performing all fundamental operations, namely matrix-vector multiplication, matrix-matrix multiplication and matrix inversion can be done in almost linear …
-
The use of the source reconstruction method for antenna characterization
… information with experimental data. Next the hierarchical matrix framework is studied in conjunction with the SRM to decrease the algorithm's memory requirement and increase the speed of execution. It is shown that it is beneficial to use the hierarchical matrix framework with the SRM when …
Page 1 of 2