# Behavioral measures and their correlation with IPM iteration counts on semi-definite programming problems

- 115 Downloads
- 10 Citations

## Abstract

We study four measures of problem instance behavior that might account for the observed differences in interior-point method (IPM) iterations when these methods are used to solve semidefinite programming (SDP) problem instances: (i) an aggregate geometry measure related to the primal and dual feasible regions (aspect ratios) and norms of the optimal solutions, (ii) the (Renegar-) condition measure *C*(*d*) of the data instance, (iii) a measure of the near-absence of strict complementarity of the optimal solution, and (iv) the level of degeneracy of the optimal solution. We compute these measures for the SDPLIB suite problem instances and measure the sample correlation (CORR) between these measures and IPM iteration counts (solved using the software SDPT3) when these measures have finite values. Our conclusions are roughly as follows: the aggregate geometry measure is highly correlated with IPM iterations (CORR = 0.901), and provides a very good explanation of IPM iterations, particularly for problem instances with solutions of small norm and aspect ratio. The condition measure *C*(*d*) is also correlated with IPM iterations, but less so than the aggregate geometry measure (CORR = 0.630). The near-absence of strict complementarity is weakly correlated with IPM iterations (CORR = 0.423). The level of degeneracy of the optimal solution is essentially uncorrelated with IPM iterations.

## Keywords

Behavioral measure Condition number Degeneracy Complementarity Interior-point method Semi-definite programming## Mathematics Subject Classification (2000)

90-04 90C22 90C60 90C51## Preview

Unable to display preview. Download preview PDF.

## References

- 1.Alizadeh F., Haeberly J.P.A., Overton M.L. (1997): Complementarity and nondegeneracy in semidefinite programming. Math. Program. 77, 111–128CrossRefMathSciNetGoogle Scholar
- 2.Alizadeh F., Haeberly J.P.A., Overton M.L. (1998): Primal-dual interior-point methods for semidefinite programming: convergence rates, stability and numerical results. SIAM J. Optim. 8(3): 746–768CrossRefMathSciNetzbMATHGoogle Scholar
- 3.Bhatia R. (1997): Matrix Analysis. Springer, Berlin Heidelberg New YorkGoogle Scholar
- 4.Cai, Z., Freund, R.M.: On two measures of problem complexity and their explanatory value for the performance of SeDuMi on second-order cone problems. Comput Optim Appl (in press)Google Scholar
- 5.Chai J., Toh K. (2006): Computation of condition numbers for linear programming problems using Peña’s method. Optim. Methods. Softw. 21, 419–443CrossRefMathSciNetzbMATHGoogle Scholar
- 6.Freund, R.M.: On the primal–dual geometry of level sets in linear and conic optimization. SIAM J. Optim.
**13**(4) (2003)Google Scholar - 7.Freund R.M. (2004): Complexity of convex optimization using geometry-based measures and a reference point. Math. Program. 99, 197–221CrossRefMathSciNetzbMATHGoogle Scholar
- 8.Freund R.M., Vera J.R. (1999): Some characterizations and properties of the “distance to ill-posedness" and the condition measure of a conic linear system. Math. Program. 86(2): 225–260CrossRefMathSciNetzbMATHGoogle Scholar
- 9.Freund R.M., Vera J.R. (2003): On the complexity of computing estimates of condition measures of a conic linear system. Math. Oper. Rese. 28(4): 625–648CrossRefMathSciNetzbMATHGoogle Scholar
- 10.Ji J., Potra F., Sheng R. (1999): On the local convergence of a predictor-corrector method for semidefinite programming. SIAM J. Optim 10(1): 195–210CrossRefMathSciNetzbMATHGoogle Scholar
- 11.Kojima M., Shida M., Shindoh S. (1998): Local convergence of predictor–corrector infeasible-interior-point algorithms for sdps and sdlcps. Math. Program. 80(2): 129–160CrossRefMathSciNetGoogle Scholar
- 12.Lu Z., Monteiro R. (2005): Error bounds and limiting behavior of weighted paths associated with the sdp map
*X*^{1/2}*SX*^{1/2}. SIAM J. Optim. 15(2): 348–374CrossRefMathSciNetGoogle Scholar - 13.Luo Z.Q., Sturm J., Zhang S. (1998): Superlinear convergence of a symmetric primal-dual path following algorithm for semidefinite programming. SIAM J. Optim. 8(1): 59–81CrossRefMathSciNetzbMATHGoogle Scholar
- 14.Ordóñez F., Freund R.M. (2003): Computational experience and the explanatory value of condition measures for linear optimization. SIAM J. Optim. 14(2): 307–333CrossRefMathSciNetzbMATHGoogle Scholar
- 15.Peña J. (2002): Two properties of condition numbers for convex programs via implicitly defined barrier functions. Math. Program. 93(1): 55–57CrossRefMathSciNetzbMATHGoogle Scholar
- 16.Potra F., Sheng R. (1998): Superlinear convergence of interior-point algorithms for semidefinite programming. J. Optim. Theory Appl. 99(1): 103–119CrossRefMathSciNetzbMATHGoogle Scholar
- 17.Potra F., Wright S. (2000): Interior-point methods. J. Comput Appl. Math. 124, 281–302CrossRefMathSciNetzbMATHGoogle Scholar
- 18.Renegar J. (1994): Some perturbation theory for linear programming. Math. Program. 65(1): 73–91CrossRefMathSciNetGoogle Scholar
- 19.Renegar J. (1995): Linear programming, complexity theory, and elementary functional analysis. Math. Program. 70(3): 279–351CrossRefMathSciNetGoogle Scholar
- 20.Robinson S.M. (1977): A characterization of stability in linear programming. Oper. Res. 25(3): 435–447CrossRefzbMATHGoogle Scholar
- 21.Toh K.C. (2004): Solving large scale semidefinite programs via an iterative solver on the augmented systems. SIAM J. Optim. 14, 670–698CrossRefMathSciNetGoogle Scholar
- 22.Tütüncü R.H., Toh K.C., Todd M.J. (2003): Solving semidefinite-quadratic-linear programs using sdpt3. Math. Program. 95(2): 189–217CrossRefMathSciNetzbMATHGoogle Scholar
- 23.Wolkowicz, H., Saigal, R., Vandenberghe, L.: Handbook of semidefinite programming. Kluwer (2000)Google Scholar