Matrix factorization ranks via polynomial optimization

In light of recent data science trends, new interest has fallen in alternative matrix factorizations. By this, we mean various ways of factorizing particular data matrices so that the factors have special properties and reveal insights into the original data. We are interested in the specialized ranks associated with these factorizations, but they are usually difficult to compute. In particular, we consider the nonnegative-, completely positive-, and separable ranks. We focus on a general tool for approximating factorization ranks, the moment hierarchy, a classical technique from polynomial optimization, further augmented by exploiting ideal-sparsity. Contrary to other examples of sparsity, the resulting sparse hierarchy yields equally strong, if not superior, bounds while potentially delivering a speed-up in computation.

[1]  Milan Korda,et al.  Exploiting ideal-sparsity in the generalized moment problem with application to matrix factorization ranks , 2022, Math. Program..

[2]  Monique Laurent,et al.  Bounding the separable rank via polynomial optimization , 2021, Linear Algebra and its Applications.

[3]  Rekha R. Thomas,et al.  Lifting for Simplicity: Concise Descriptions of Convex Sets , 2020, SIAM Rev..

[4]  Tim Netzer,et al.  Mixed states in one spatial dimension: Decompositions and correspondence with nonnegative matrices , 2019, Journal of Mathematical Physics.

[5]  Tim Netzer,et al.  Separability for mixed states with operator Schmidt rank two , 2019, Quantum.

[6]  E. D. Klerk,et al.  A Survey of Semidefinite Programming Approaches to the Generalized Problem of Moments and Their Error Analysis , 2018, Association for Women in Mathematics Series.

[7]  Joel A. Tropp,et al.  Nonnegative Matrix Factorization , 2018 .

[8]  Monique Laurent,et al.  Lower Bounds on Matrix Factorization Ranks via Noncommutative Polynomial Optimization , 2017, Found. Comput. Math..

[9]  Iain Dunning,et al.  JuMP: A Modeling Language for Mathematical Optimization , 2015, SIAM Rev..

[10]  Immanuel M. Bomze,et al.  New Lower Bounds and Asymptotics for the cp-Rank , 2015, SIAM J. Matrix Anal. Appl..

[11]  Alan Edelman,et al.  Julia: A Fresh Approach to Numerical Computing , 2014, SIAM Rev..

[12]  Immanuel M. Bomze,et al.  From seven to eleven: Completely positive matrices with high cp-rank , 2014 .

[13]  Rekha R. Thomas,et al.  Positive semidefinite rank , 2014, Math. Program..

[14]  Pablo A. Parrilo,et al.  Self-scaled bounds for atomic cone ranks: applications to nonnegative rank and cp-rank , 2014, Math. Program..

[15]  Peter J. C. Dickinson,et al.  On the computational complexity of membership problems for the completely positive cone and its dual , 2014, Comput. Optim. Appl..

[16]  Florian Jarre,et al.  On the cp-Rank and Minimal cp Factorizations of a Completely Positive Matrix , 2013, SIAM J. Matrix Anal. Appl..

[17]  Jiawang Nie,et al.  The A-Truncated K -Moment Problem , 2012 .

[18]  Hans Raj Tiwary,et al.  Exponential Lower Bounds for Polytopes in Combinatorial Optimization , 2011, J. ACM.

[19]  Nicolas Gillis,et al.  On the Geometric Interpretation of the Nonnegative Rank , 2010, 1009.0880.

[20]  J. Lasserre Moments, Positive Polynomials And Their Applications , 2009 .

[21]  S. Amari,et al.  Nonnegative Matrix and Tensor Factorizations - Applications to Exploratory Multi-way Data Analysis and Blind Source Separation , 2009 .

[22]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[23]  Samuel Burer,et al.  On the copositive representation of binary and continuous nonconvex quadratic programs , 2009, Math. Program..

[24]  Sevag Gharibian,et al.  Strong NP-hardness of the quantum separability problem , 2008, Quantum Inf. Comput..

[25]  Stephen A. Vavasis,et al.  On the Complexity of Nonnegative Matrix Factorization , 2007, SIAM J. Optim..

[26]  L. Gurvits Classical deterministic complexity of Edmonds' Problem and quantum entanglement , 2003, STOC '03.

[27]  N. Sidiropoulos,et al.  On the uniqueness of multilinear decomposition of N‐way arrays , 2000 .

[28]  R. Curto,et al.  The truncated complex -moment problem , 2000 .

[29]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[30]  Ashish V. Thapliyal,et al.  Optimal decompositions of barely separable states , 1999, quant-ph/9904005.

[31]  A. Uhlmann Entropy and Optimal Decompositions of States Relative to a Maximal Commutative Subalgebra , 1997, quant-ph/9704017.

[32]  Charles R. Johnson,et al.  Completely positive matrices associated with M-matrices , 1994 .

[33]  Mihalis Yannakakis,et al.  Expressing combinatorial optimization problems by linear programs , 1991, STOC '88.

[34]  C. Bron,et al.  Algorithm 457: finding all cliques of an undirected graph , 1973 .

[35]  di Dio,et al.  The truncated moment problem , 2018 .

[36]  M. Laurent Sums of Squares, Moment Matrices and Optimization Over Polynomials , 2009 .

[37]  J. Lasserre,et al.  Detecting global optimality and extracting solutions in GloptiPoly , 2003 .

[38]  Knud D. Andersen,et al.  The Mosek Interior Point Optimizer for Linear Programming: An Implementation of the Homogeneous Algorithm , 2000 .

[39]  A. Fialkow,et al.  THE TRUNCATED COMPLEX K-MOMENT PROBLEM , 2000 .

[40]  Raúl E. Curto,et al.  Solution of the Truncated Complex Moment Problem for Flat Data , 1996 .

[41]  Yurii Nesterov,et al.  Interior-point polynomial algorithms in convex programming , 1994, Siam studies in applied mathematics.

[42]  安藤 毅 Completely positive matrices , 1991 .

[43]  A. Berman,et al.  Copositive and Completely Positive Matrices , 2022 .