Sparse sum-of-squares (SOS) optimization: A bridge between DSOS/SDSOS and SOS optimization for sparse polynomials

Optimization over non-negative polynomials is fundamental for nonlinear systems analysis and control. This work investigates the relation between three tractable relaxations for optimizing over sparse non-negative polynomials: sparse sum-of-squares (SSOS) optimization, diagonally dominant sum-of-squares (DSOS) optimization, and scaled diagonally dominant sum-of-squares (SDSOS) optimization. We prove that the set of SSOS polynomials, an inner approximation of the cone of SOS polynomials, strictly contains the spaces of sparse DSOS/SDSOS polynomials. For problems with sparse polynomials, therefore, SSOS optimization is less conservative than its DSOS/SDSOS counterparts. Numerical results for large-scale sparse polynomial optimization problems demonstrate this fact, and also that SSOS optimization can be faster than DSOS/SDSOS methods despite requiring the solution of semidefinite programs instead of less expensive linear/second-order cone programs.

[1]  Antonis Papachristodoulou,et al.  Positive Forms and Stability of Linear Time-Delay Systems , 2006, CDC.

[2]  Amir Ali Ahmadi,et al.  Optimization over structured subsets of positive semidefinite matrices via column generation , 2015, Discret. Optim..

[3]  P. Parrilo,et al.  Symmetry groups, semidefinite programs, and sums of squares , 2002, math/0211450.

[4]  Jean B. Lasserre,et al.  Sparse-BSOS: a bounded degree SOS hierarchy for large scale polynomial optimization with sparsity , 2016, Mathematical Programming Computation.

[5]  Andreas Griewank,et al.  On the existence of convex decompositions of partially separable functions , 1984, Math. Program..

[6]  Pablo A. Parrilo,et al.  Semidefinite programming relaxations for semialgebraic problems , 2003, Math. Program..

[7]  Daniel K. Molzahn,et al.  Lasserre Hierarchy for Large Scale Polynomial Optimization in Real and Complex Variables , 2017, SIAM J. Optim..

[8]  L. Rodman,et al.  Positive semidefinite matrices with a given sparsity pattern , 1988 .

[9]  A. Papachristodoulou,et al.  Polynomial sum of squares in fluid dynamics: a review with a look ahead , 2014, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[10]  James Demmel,et al.  Sparse SOS Relaxations for Minimizing Functions that are Summations of Small Polynomials , 2008, SIAM J. Optim..

[11]  M. Kojima.  Sums of Squares Relaxations of Polynomial Semidefinite Programs , 2003 .

[12]  Carsten W. Scherer,et al.  Matrix Sum-of-Squares Relaxations for Robust Semi-Definite Programs , 2006, Math. Program..

[13]  Deqing Huang,et al.  Bounds for Deterministic and Stochastic Dynamical Systems using Sum-of-Squares Optimization , 2015, SIAM J. Appl. Dyn. Syst..

[14]  Pablo A. Parrilo,et al.  Basis selection for SOS programs via facial reduction and polyhedral approximations , 2014, 53rd IEEE Conference on Decision and Control.

[15]  Jean B. Lasserre,et al.  Global Optimization with Polynomials and the Problem of Moments , 2000, SIAM J. Optim..

[16]  Yang Zheng,et al.  Decomposition and Completion of Sum-of-Squares Matrices , 2018, 2018 IEEE Conference on Decision and Control (CDC).

[17]  A. Packard,et al.  Stability region analysis using sum of squares programming , 2006, 2006 American Control Conference.

[18]  B. Reznick.  Extremal PSD forms with few terms , 1978 .

[19]  Martin S. Andersen,et al.  Chordal Graphs and Semidefinite Optimization , 2015, Found. Trends Optim..

[20]  Johan Löfberg,et al.  YALMIP : a toolbox for modeling and optimization in MATLAB , 2004 .

[21]  Masakazu Muramatsu,et al.  Sums of Squares and Semidefinite Programming Relaxations for Polynomial Optimization Problems with Structured Sparsity , 2004 .

[22]  Maryam Kamgarpour,et al.  Scalable analysis of linear networked systems via chordal decomposition , 2018, 2018 European Control Conference (ECC).

[23]  Amir Ali Ahmadi,et al.  Sum of Squares Basis Pursuit with Linear and Second Order Cone Programming , 2015, ArXiv.

[24]  Amir Ali Ahmadi,et al.  DSOS and SDSOS Optimization: More Tractable Alternatives to Sum of Squares and Semidefinite Optimization , 2017, SIAM J. Appl. Algebra Geom..

[25]  A. Papachristodoulou,et al.  On the construction of Lyapunov functions using the sum of squares decomposition , 2002, Proceedings of the 41st IEEE Conference on Decision and Control, 2002..

[26]  Mirjam Dür,et al.  Copositive Programming – a Survey , 2010 .

[27]  Johan Löfberg,et al.  Pre- and Post-Processing Sum-of-Squares Programs in Practice , 2009, IEEE Transactions on Automatic Control.

[28]  David Goluskin,et al.  Bounding Averages Rigorously Using Semidefinite Programming: Mean Moments of the Lorenz System , 2016, J. Nonlinear Sci..