Skip to main content
Log in

Some upper and lower bounds on PSD-rank

  • Full Length Paper
  • Series A
  • Published:
Mathematical Programming Submit manuscript

Abstract

Positive semidefinite rank (PSD-rank) is a relatively new complexity measure on matrices, with applications to combinatorial optimization and communication complexity. We first study several basic properties of PSD-rank, and then develop new techniques for showing lower bounds on the PSD-rank. All of these bounds are based on viewing a positive semidefinite factorization of a matrix M as a quantum communication protocol. These lower bounds depend on the entries of the matrix and not only on its support (the zero/nonzero pattern), overcoming a limitation of some previous techniques. We compare these new lower bounds with known bounds, and give examples where the new ones are better. As an application we determine the PSD-rank of (approximations of) some common matrices.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Notes

  1. \(\mathrm {Tr}_B(\rho \otimes \sigma )=\mathrm {Tr}(\sigma )\rho \), which is extended linearly to states that are not tensor products.

  2. Even though a nonnegative matrix has the same PSD-rank as its transposition, the bounds given by \(B_3\) (or \(B_4\)) can be quite different, for instance for the matrix A of Example 2.

  3. A Latin square is an n-by-n matrix in which each row and each column is a permutation of \(\{0,\ldots ,n-1\}\).

References

  1. Gouveia, J., Parrilo, P., Thomas, R.: Lifts of convex sets and cone factorizations. Math. Oper. Res. 38(2), 248–264 (2013). arXiv:1111.3164

    Article  MathSciNet  MATH  Google Scholar 

  2. Fiorini, S., Massar, S., Pokutta, S., Tiwary, H. R. and de Wolf, R.: Exponential lower bounds for polytopes in combinatorial optimization. J. ACM, 16(2), 2015. Earlier version in STOC’12. arXiv:1111.0837

  3. Yannakakis, M.: Expressing combinatorial optimization problems by linear programs. J. Comput. Syst. Sci. 43(3), 441–466 (1991)

    Article  MathSciNet  MATH  Google Scholar 

  4. Rothvoß, T.: The matching polytope has exponential extension complexity. In: Proceedings of 46th ACM STOC, pp. 263–272 (2014)

  5. Lee, J., Steurer, D. and Raghavendra, P.: Lower bounds on the size of semidefinite programming relaxations. In: Proceedings of 47th ACM STOC, pp. 567–576 (2015). arXiv:1411.6317

  6. Lee, T. and Wei, Z.: The square root rank of the correlation polytope is exponential (2014). arXiv:1411.6712

  7. Conforti, M., Faenza, Y., Fiorini, S. and Tiwary, H.R.: Extended formulations, non-negative factorizations and randomized communication protocols. In: 2nd International Symposium on Combinatorial Optimization, pp. 129–140 (2012). arXiv:1105.4127

  8. Zhang, S.: Quantum strategic game theory. In: Proceedings of the 3rd Innovations in Theoretical Computer Science, pp 39–59 (2012). arXiv:1012.5141

  9. Jain, R., Shi, Y., Wei, Z., Zhang, S.: Efficient protocols for generating bipartite classical distributions and quantum states. IEEE Trans. Inf. Theory 59, 5171–5178 (2013)

    Article  MathSciNet  Google Scholar 

  10. Kushilevitz, E., Nisan, N.: Communication Complexity. Cambridge University Press, Cambridge (1997)

    Book  MATH  Google Scholar 

  11. Lee, T. and Theis, D.O.: Support based bounds for positive semidefinite rank (2012). arXiv:1203.3961

  12. Sikora, J., Varvitsiotis, A. and Wei, Z.: On the minimum dimension of a Hilbert space needed to generate a quantum correlation (2015). arXiv:1507.00213

  13. Nielsen, M., Chuang, I.: Quantum Computation and Quantum Information. Cambridge University Press, Cambridge (2000)

    MATH  Google Scholar 

  14. Ben-Tal, A., Nemirovski, A.: On polyhedral approximations of the second-order cone. Math. Oper. Res. 26(2), 193–205 (2001)

    Article  MathSciNet  MATH  Google Scholar 

  15. Fawzi, H., Parrilo, P.: Lower bounds on nonnegative rank via nonnegative nuclear norms. Math. Program., Ser. B 153(1), 41–66 (2015). arXiv:1210.6970

    Article  MathSciNet  MATH  Google Scholar 

  16. Fawzi, H., Gouveia, J., Parrilo, P., Robinson, R., Thomas, R.: Positive semidefinite rank. Math. Program., Ser. B 153(1), 133–177 (2015). arXiv:1407.4095

    Article  MathSciNet  MATH  Google Scholar 

  17. Alon, N.: Perturbed identity matrices have high rank: proof and applications. Combin., Prob., Comput. 18, 3–15 (2009)

    Article  MathSciNet  MATH  Google Scholar 

Download references

Acknowledgments

We would like to thank Rahul Jain for helpful discussions, and Hamza Fawzi, Richard Robinson, and Rekha Thomas for sharing their results on the derangement matrix. Troy Lee and Zhaohui Wei are supported in part by the Singapore National Research Foundation under NRF RF Award No. NRF-NRFF2013-13. Ronald de Wolf is partially supported by ERC Consolidator Grant QPROGRESS and by the EU STREP project QALGO (Grant Agreement No. 600700).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Zhaohui Wei.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lee, T., Wei, Z. & de Wolf, R. Some upper and lower bounds on PSD-rank. Math. Program. 162, 495–521 (2017). https://doi.org/10.1007/s10107-016-1052-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10107-016-1052-0

Keywords

Mathematics Subject Classification

Navigation