Compressing Rank-Structured Matrices via Randomized Sampling

Randomized sampling has recently been proven a highly efficient technique for computing approximate factorizations of matrices that have low numerical rank. This paper describes an extension of such techniques to a wider class of matrices that are not themselves rank-deficient, but have off-diagonal blocks that are; specifically, the classes of so called \textit{Hierarchically Off-Diagonal Low Rank (HODLR)} matrices and \textit{Hierarchically Block Separable (HBS)} matrices. Such matrices arise frequently in numerical analysis and signal processing, in particular in the construction of fast methods for solving differential and integral equations numerically. These structures admit algebraic operations (matrix-vector multiplications, matrix factorizations, matrix inversion, etc.) to be performed very rapidly; but only once a data-sparse representation of the matrix has been constructed. How to rapidly compute this representation in the first place is much less well understood. The present paper demonstrates that if an $N\times N$ matrix can be applied to a vector in $O(N)$ time, and if the ranks of the off-diagonal blocks are bounded by an integer $k$, then the cost for constructing a HODLR representation is $O(k^{2}\,N\,(\log N)^{2})$, and the cost for constructing an HBS representation is $O(k^{2}\,N\,\log N)$. The point is that when legacy codes (based on, e.g., the Fast Multipole Method) can be used for the fast matrix-vector multiply, the proposed algorithm can be used to obtain the data-sparse representation of the matrix, and then well-established techniques for HODLR/HBS matrices can be used to invert or factor the matrix. The proposed scheme is also useful in simplifying the implementation of certain operations on rank-structured matrices such as the matrix-matrix multiplication, low-rank update, addition, etc.

[1]  Robert H. Halstead,et al.  Matrix Computations , 2011, Encyclopedia of Parallel Computing.

[2]  C. Chui,et al.  Article in Press Applied and Computational Harmonic Analysis a Randomized Algorithm for the Decomposition of Matrices , 2022 .

[3]  Jianlin Xia,et al.  Randomized Sparse Direct Solvers , 2013, SIAM J. Matrix Anal. Appl..

[4]  Faculteit Ingenieurswetenschappen,et al.  MULTISCALE AND HYBRID METHODS FOR THE SOLUTION OF OSCILLATORY INTEGRAL EQUATIONS , 2006 .

[5]  Jianlin Xia,et al.  Superfast Multifrontal Method for Large Structured Linear Systems of Equations , 2009, SIAM J. Matrix Anal. Appl..

[6]  I. Duff,et al.  Direct Methods for Sparse Matrices , 1987 .

[7]  W. Hackbusch.  A Sparse Matrix Arithmetic Based on $\Cal H$-Matrices. Part I: Introduction to ${\Cal H}$-Matrices , 1999, Computing.

[8]  Per-Gunnar Martinsson,et al.  A Fast Randomized Algorithm for Computing a Hierarchically Semiseparable Representation of a Matrix , 2011, SIAM J. Matrix Anal. Appl..

[9]  Timothy A. Davis,et al.  Direct methods for sparse linear systems , 2006, Fundamentals of algorithms.

[10]  V. Rokhlin,et al.  A randomized algorithm for the approximation of matrices , 2006 .

[11]  Per-Gunnar Martinsson,et al.  A Fast Direct Solver for a Class of Elliptic Partial Differential Equations , 2009, J. Sci. Comput..

[12]  Steffen Börm,et al.  Hybrid cross approximation of integral operators , 2005, Numerische Mathematik.

[13]  Ming Gu,et al.  Efficient Algorithms for Computing a Strong Rank-Revealing QR Factorization , 1996, SIAM J. Sci. Comput..

[14]  W. Hackbusch,et al.  On H2-Matrices , 2000 .

[15]  Marc Van Barel,et al.  Solving a large dense linear system by adaptive cross approximation , 2010, J. Comput. Appl. Math..

[16]  Lexing Ying,et al.  Fast construction of hierarchical matrix representation from matrix-vector multiplication , 2009, J. Comput. Phys..

[17]  L. Grasedyck,et al.  Domain-decomposition Based ℌ-LU Preconditioners , 2007 .

[18]  V. Rokhlin,et al.  A fast direct solver for boundary integral equations in two dimensions , 2003 .

[19]  Sjsu ScholarWorks,et al.  Rank revealing QR factorizations , 2014 .

[20]  Per-Gunnar Martinsson,et al.  A direct solver with O(N) complexity for integral equations on one-dimensional domains , 2011, 1105.5372.

[21]  Wolfgang Hackbusch,et al.  A Sparse Matrix Arithmetic Based on H-Matrices. Part I: Introduction to H-Matrices , 1999, Computing.

[22]  A. George.  Nested Dissection of a Regular Finite Element Mesh , 1973 .

[23]  Steffen Börm,et al.  Data-sparse Approximation by Adaptive ℋ2-Matrices , 2002, Computing.

[24]  Per-Gunnar Martinsson,et al.  An Accelerated Kernel-Independent Fast Multipole Method in One Dimension , 2007, SIAM J. Sci. Comput..

[25]  S. Börm.  Efficient Numerical Methods for Non-local Operators , 2010 .

[26]  Mario Bebendorf,et al.  Approximation of boundary element matrices , 2000, Numerische Mathematik.

[27]  Nathan Halko,et al.  Finding Structure with Randomness: Probabilistic Algorithms for Constructing Approximate Matrix Decompositions , 2009, SIAM Rev..

[28]  Per-Gunnar Martinsson,et al.  A Direct Solver with O(N) Complexity for Variable Coefficient Elliptic PDEs Discretized via a High-Order Composite Spectral Collocation Method , 2013, SIAM J. Sci. Comput..

[29]  Timothy A. Davis,et al.  Algorithm 832: UMFPACK V4.3---an unsymmetric-pattern multifrontal method , 2004, TOMS.

[30]  Leslie Greengard,et al.  A fast algorithm for particle simulations , 1987 .

[31]  Per-Gunnar Martinsson,et al.  On the Compression of Low Rank Matrices , 2005, SIAM J. Sci. Comput..

[32]  P. Martinsson.  Rapid factorization of structured matrices via randomized sampling , 2008, 0806.2339.

[33]  Jianlin Xia,et al.  Fast algorithms for hierarchically semiseparable matrices , 2010, Numer. Linear Algebra Appl..

[34]  S. Chandrasekaran,et al.  A fast adaptive solver for hierarchically semiseparable representations , 2005 .