ISAAC 2013: Algorithms and Computation pp 119-129

# Determinantal Complexities and Field Extensions

• Youming Qiao
• Xiaoming Sun
• Nengkun Yu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8283)

## Abstract

Let $$\mathbb{F}$$ be a field of characteristic ≠ 2. The determinantal complexity of a polynomial $$P\in\mathbb{F}[x_1, \dots, x_n]$$ is defined as the smallest size of a matrix M whose entries are linear polynomials of x i ’s over $$\mathbb{F}$$, such that $$P=\det (M)$$ as polynomials in $$\mathbb{F}[x_1, \dots, x_n]$$. To determine the determinantal complexity of the permanent polynomial is a long-standing open problem.

Let $$\mathbb{K}$$ be an extension field of $$\mathbb{F}$$; then P can be viewed as a polynomial over $$\mathbb{K}$$. We are interested in the comparison between the determinantal complexity of P over $$\mathbb{K}$$ (denoted as $$\mathtt{dc}_{\mathbb{K}}(P)$$), and that of P over $$\mathbb{F}$$ (denoted as $$\mathtt{dc}_{\mathbb{F}}(P)$$). It is clear that $$\mathtt{dc}_{\mathbb{K}}(P)\leq \mathtt{dc}_\mathbb{F}(P)$$, and the question is whether strict inequality can happen. In this note we consider polynomials defined over ℚ. For $$P=x_1^2+\dots+x_n^2$$, there exists a constant multiplicative gap between dc (P) and dc (P): we prove dc (P) ≥ n while ⌈n/2⌉ + 1 ≥ dc (P). We also consider additive constant gaps: (1) there exists a quadratic polynomial Q ∈ ℚ[x, y], such that dc (Q) = 3 and $$\mathtt{dc}_{\overline{\mathbb{Q}}}(Q)=2$$; (2) there exists a cubic polynomial C ∈ ℚ[x, y] with a rational zero, such that dc (C) = 4 and $$\mathtt{dc}_{\overline{\mathbb{Q}}}(C)=3$$. For additive constant gaps, geometric criteria are presented to decide when $$\mathtt{dc}_{\mathbb{Q}}=\mathtt{dc}_{\overline{\mathbb{Q}}}$$.

## Preview

### References

1. 1.
Agrawal, M.: Determinant versus permanent. In: Proceedings of the International Congress of Mathematicians (ICM), pp. 1409–1421 (2006)Google Scholar
2. 2.
Bürgisser, P.: Completeness and Reduction in Algebraic Complexity Theory. In: Algorithms and Computation in Mathematics. Springer (2000)Google Scholar
3. 3.
Bürgisser, P., Clausen, M., Shokrollahi, M.A.: Algebraic Complexity Theory. A series of comprehensive studies in mathematics. Springer (1997)Google Scholar
4. 4.
Cai, J.-Y.: A note on the determinant and permanent problem. Information and Computation 84(1), 119–127 (1990)
5. 5.
Cai, J.-Y., Chen, X., Li, D.: A quadratic lower bound for the permanent and determinant problem over any characteristic ≠ 2. In: Proceedings of the 40th Annual ACM Symposium on Theory of Computing, pp. 491–498 (2008)Google Scholar
6. 6.
Dickson, L.E.: Algebraic theory of the expressibility of cubic forms as determinants, with application to diophantine analysis. American Journal of Mathematics 43(2), 102–125 (1921)
7. 7.
Dickson, L.E.: Determination of all general homogeneous polynomials expressible as determinants with linear elements. Transactions of the American Mathematical Society 22(2), 167–179 (1921)
8. 8.
Hrubes, P., Wigderson, A., Yehudayoff, A.: Relationless completeness and separations. In: IEEE Conference on Computational Complexity, pp. 280–290 (2010)Google Scholar
9. 9.
Hrubes, P., Yehudayoff, A.: Arithmetic complexity in ring extensions. Theory of Computing 7(1), 119–129 (2011)
10. 10.
Malod, G., Portier, N.: Characterizing valiant’s algebraic complexity classes. J. Complex. 24(1), 16–38 (2008)
11. 11.
Mignon, T., Ressayre, N.: A quadratic bound for the determinant and permanent problem. In: International Mathematics Research Notices, pp. 4241–4253 (2004)Google Scholar
12. 12.
Strassen, V.: Berechnung und programm. I. Acta Inf. 1, 320–335 (1972)
13. 13.
Toda, S.: Classes of arithmetic circuits capturing the complexity of computing the determinant. IEICE Trans. Inf. Syst. E75-D, 116–124 (1992)Google Scholar
14. 14.
Valiant, L.G.: Completeness classes in algebra. In: Proceedings of the Eleventh Annual ACM Symposium on Theory of Computing, STOC 1979, pp. 249–261. ACM, New York (1979)
15. 15.
von, J., von zur Gathen, J.: Permanent and determinant. Linear Algebra and its Applications 96, 87–100 (1987)

## Authors and Affiliations

• Youming Qiao
• 1
• Xiaoming Sun
• 2
• Nengkun Yu
• 3
• 4
1. 1.Centre for Quantum TechnologiesThe National University of SingaporeSingapore
2. 2.Institute of Computing TechnologyChinese Academy of SciencesChina
3. 3.State Key Laboratory of Intelligent Technology and Systems, Tsinghua National Laboratory for Information Science and Technology, Department of Computer Science and TechnologyTsinghua UniversityBeijingChina
4. 4.Centre for Quantum Computation and Intelligent Systems (QCIS), Faculty of Engineering and Information TechnologyUniversity of TechnologySydneyAustralia