Investigation and Implementation of Matrix Permanent Algorithms for Identity Resolution

Abstract : Identification of tracks/targets forms a cornerstone of Maritime Situational Awareness (MSA) in both tactical and operational settings. Resolving the identities of unknown targets often demands significant resources, and thus it is highly desirable to improve automatic vessel identification to the greatest extent that is computationally possible. Recent work at Defence Research and Development Canada (DRDC), Atlantic Research Centre has studied the statistical identification problem and found that it depends crucially on calculation of the permanent of a matrix whose dimension is a function of target count. However, the optimal approach for computing the permanent is presently unclear. The primary objective of this project was to determine the optimal computing strategy(-ies) for the matrix permanent in tactical and operational scenarios. As the exact calculation is #P-hard, approximation methods are necessary. However, many new and promising methods lack characterization, particularly in regard to their suitability for MSA applications. The present work seeks to clarify the computational options available and includes the following elements: (1.) Identification and characterization of algorithms described in the scientific literature. (2.) Implementation of those algorithms most suited for MSA.

[1]  Richard J. Lipton,et al.  A Monte-Carlo Algorithm for Estimating the Permanent , 1993, SIAM J. Comput..

[2]  Yan Zhou.  vSMC: Parallel Sequential Monte Carlo in C++ , 2013, ArXiv.

[3]  Eric Vigoda,et al.  A polynomial-time approximation algorithm for the permanent of a matrix with non-negative entries , 2001, STOC '01.

[4]  Eric Vigoda,et al.  A polynomial-time approximation algorithm for the permanent of a matrix with nonnegative entries , 2004, JACM.

[5]  Mark Jerrum,et al.  Approximating the Permanent , 1989, SIAM J. Comput..

[6]  P. W. Kasteleyn.  The statistics of dimers on a lattice: I. The number of dimer arrangements on a quadratic lattice , 1961 .

[7]  I. Beichl,et al.  Approximating the Permanent via Importance Sampling with Application to the Dimer Covering Problem , 1999 .

[8]  David G. Glynn,et al.  The permanent of a square matrix , 2010, Eur. J. Comb..

[9]  Michael Chertkov,et al.  Approximating the permanent with fractional belief propagation , 2011, J. Mach. Learn. Res..

[10]  V. Vazirani,et al.  Accelerating simulated annealing for the permanent and combinatorial counting problems , 2006, SODA 2006.

[11]  Donald E. Knuth,et al.  The Art of Computer Programming, Volume 4, Fascicle 2: Generating All Tuples and Permutations (Art of Computer Programming) , 2005 .

[12]  Jun S. Liu,et al.  SEQUENTIAL MONTE CARLO METHODS FOR PERMUTATION TESTS ON TRUNCATED DATA , 2007 .

[13]  Jeffrey K. Uhlmann,et al.  Matrix permanent inequalities for approximating joint assignment matrices in tracking systems , 2004, J. Frankl. Inst..

[14]  N. Y. Kuznetsov.  Computing the permanent by importance sampling method , 1996 .

[15]  Alexander I. Barvinok,et al.  Polynomial Time Algorithms to Approximate Permanents and Mixed Discriminants Within a Simply Exponential Factor , 1999, Random Struct. Algorithms.

[16]  Leslie G. Valiant,et al.  The Complexity of Computing the Permanent , 1979, Theor. Comput. Sci..

[17]  Leslie G. Valiant,et al.  Random Generation of Combinatorial Structures from a Uniform Distribution , 1986, Theor. Comput. Sci..

[18]  Mark R. Morelande.  Joint data association using importance sampling , 2009, 2009 12th International Conference on Information Fusion.

[19]  Bert Huang,et al.  Approximating the Permanent with Belief Propagation , 2009, ArXiv.

[20]  Andrei Z. Broder,et al.  How hard is it to marry at random? (On the approximation of the permanent) , 1986, STOC '86.

[21]  Brian P. Dawkins,et al.  Estimating the permanent by importance sampling from a finite population , 2001 .