Kernelization of Constraint Satisfaction Problems: A Study Through Universal Algebra
A kernelization algorithm for a computational problem is a procedure which compresses an instance into an equivalent instance whose size is bounded with respect to a complexity parameter. For the constraint satisfaction problem (CSP), there exist many results concerning upper and lower bounds for kernelizability of specific problems, but it is safe to say that we lack general methods to determine whether a given problem admits a kernel of a particular size. In this paper, we take an algebraic approach to the problem of characterizing the kernelization limits of NP-hard CSP problems, parameterized by the number of variables. Our main focus is on problems admitting linear kernels, as has, somewhat surprisingly, previously been shown to exist. We show that a finite-domain CSP problem has a kernel with O(n) constraints if it can be embedded (via a domain extension) into a CSP which is preserved by a Maltsev operation. This result utilise a variant of the simple algorithm for Maltsev constraints. In the complementary direction, we give indication that the Maltsev condition might be a complete characterization for Boolean CSPs with linear kernels, by showing that an algebraic condition that is shared by all problems with a Maltsev embedding is also necessary for the existence of a linear kernel unless NP \(\subseteq \) co-NP/poly.
We thank the anonymous reviewers for several helpful suggestions. The first author is supported by the DFG-funded project “Homogene Strukturen, Bedingungserfüllungsprobleme, und topologische Klone” (Project number 622397).
- 1.Bulatov, A.: A dichotomy theorem for nonuniform CSPs. CoRR, abs/1703.03021 (2017)Google Scholar
- 11.Jansen, B.M.P., Pieterse, A.: Sparsification upper and lower bounds for graphs problems and not-all-equal SAT. In: Proceedings of IPEC 2015, Patras, Greece (2015)Google Scholar
- 12.Jansen, B.M.P., Pieterse, A.: Optimal sparsification for some binary CSPs using low-degree polynomials. In: Proceedings of MFCS 2016, vol. 58, pp. 71:1–71:14 (2016)Google Scholar
- 20.Lagerkvist, V., Wahlström, M.: The power of primitive positive definitions with polynomially many variables. JLC 27, 1465–1488 (2016)Google Scholar
- 21.Lagerkvist, V., Wahlström, M.: Kernelization of constraint satisfaction problems: a study through universal algebra. ArXiv e-prints, June 2017Google Scholar
- 22.Lagerkvist, V., Wahlström, M., Zanuttini, B.: Bounded bases of strong partial clones. In: Proceedings of ISMVL 2015 (2015)Google Scholar
- 26.Rafiey, A., Kinne, J., Feder, T.: Dichotomy for digraph homomorphism problems. CoRR, abs/1701.02409 (2017)Google Scholar
- 27.Zhuk, D.: The proof of CSP dichotomy conjecture. CoRR, abs/1704.01914 (2017)Google Scholar