Pattern graph for sparse Hessian matrix determination†

In a recent work, we have proposed the pattern graph as a unifying framework for methods that exploit sparsity by matrix compression: row compression, column compression or a combination of the two in sparse Jacobian matrix determination. Utilization of structural similarity between the matrix and its graph has been found to be beneficial. In this paper, we show that an important structural property, symmetry, can be exploited in the formulation of sparse Hessian matrix calculations using the pattern graph model. Using the notion of ‘direct cover’, we present a new general direct method for the determination of sparse Hessian matrices with fixed sparsity pattern and a multicolouring interpretation of it on the pattern graph associated with the matrix. A heuristic procedure for finding direct covers is sketched and some preliminary numerical test results are provided.

[1]  A. Griewank,et al.  Efficient Evaluation of Sparse Jacobians by Matrix Compression , Part I : Motivation and Theory , 2014 .

[2]  Alex Pothen,et al.  New Acyclic and Star Coloring Algorithms with Application to Computing Hessians , 2007, SIAM J. Sci. Comput..

[3]  John D. Ramsdell,et al.  Estimation of Sparse Jacobian Matrices , 1983 .

[4]  A. Griewank,et al.  Efficient Evaluation of Sparse Jacobians by Matrix Compression Part II: Implementation and Experiments , 2014 .

[5]  Shahadat Hossain,et al.  Optimal direct determination of sparse Jacobian matrices , 2013, Optim. Methods Softw..

[6]  Andrea Walther,et al.  Efficient Computation of Sparse Hessians Using Coloring and Automatic Differentiation , 2009, INFORMS J. Comput..

[7]  Shahadat Hossain,et al.  DSJM: A Software Toolkit for Direct Determination of Sparse Jacobian Matrices , 2016, ICMS.

[8]  Andreas Griewank,et al.  Evaluating derivatives - principles and techniques of algorithmic differentiation, Second Edition , 2000, Frontiers in applied mathematics.

[9]  S. Hossain,et al.  Determining Sparse Jacobian Matrices Using Two-Sided Compression: An Algorithm and Lower Bounds , 2016 .

[10]  M. Powell,et al.  On the Estimation of Sparse Hessian Matrices , 1979 .

[11]  D. Gay,et al.  More AD of Nonlinear AMPL Models: Computing Hessian Information and Exploiting Partial Separability † , 1996 .

[12]  Shahadat Hossain,et al.  Computing Sparse Jacobian Matrices Optimally , 2006 .

[13]  Frank Thomson Leighton,et al.  A Graph Coloring Algorithm for Large Scheduling Problems. , 1979, Journal of research of the National Bureau of Standards.

[14]  Alex Pothen,et al.  What Color Is Your Jacobian? Graph Coloring for Computing Derivatives , 2005, SIAM Rev..

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

[16]  T. Steihaug,et al.  Computing a sparse Jacobian matrix by rows and columns , 1998 .

[17]  Alex Pothen,et al.  ColPack: Software for graph coloring and related problems in scientific computing , 2013, TOMS.

[18]  Shahadat Hossain,et al.  Graph models and their efficient implementation for sparse Jacobian matrix determination , 2013, Discret. Appl. Math..

[19]  Thomas F. Coleman,et al.  Estimation of sparse hessian matrices and graph coloring problems , 1982, Math. Program..

[20]  S. Thomas McCormick,et al.  Optimal approximation of sparse hessians and its equivalence to a graph coloring problem , 1983, Math. Program..