Frontiers in Algorithmics and Algorithmic Aspects in Information and Management pp 192-198 | Cite as
Some Remarks on the Incompressibility of Width-Parameterized SAT Instances
Abstract
Compressibility of a formula regards reducing the length of the input, or some other parameter, while preserving the solution. Any 3-SAT instance on N variables can be represented by O(N 3) bits; [4] proved that the instance length in general cannot be compressed to O(N 3 − ε ) bits under the assumption \(\mathbf{NP}\not\subseteq\mathbf{coNP}\) /poly, which implies that the polynomial hierarchy does not collapse. This note initiates research on compressibility of SAT instances parameterized by width parameters, such as tree-width or path-width. Let SAT tw (w(n)) be the satisfiability instances of length n that are given together with a tree-decomposition of width O(w(n)), and similarly let SAT pw (w(n)) be instances with a path-decomposition of width O(w(n)). Applying simple techniques and observations, we prove conditional incompressibility for both instance length and width parameters: (i) under the exponential time hypothesis, given an instance φ of SAT tw (w(n)) it is impossible to find within polynomial time a φ′ that is satisfiable if and only if φ is satisfiable and tree-width of φ′ is half of φ; and (ii) assuming a scaled version of \(\mathbf{NP}\not\subseteq\mathbf{coNP}\) /poly, any 3-SAT pw (w(n)) instance of N variables cannot be compressed to O(N 1 − ε ) bits.
Keywords
Conjunctive Normal Form Tree Decomposition Width Parameter Input Instance Input LengthPreview
Unable to display preview. Download preview PDF.
References
- 1.Alekhnovich, M., Razborov, A.A.: Satisfiability, branch-width and Tseitin tautologies. In: Foundations of Computer Science (FOCS), pp. 593–603. IEEE (2002)Google Scholar
- 2.Allender, E., Chen, S., Lou, T., Papakonstantinou, P., Tang, B.: Width-parameterized sat: Time-space tradeoffs (2012) (manuscript)Google Scholar
- 3.Bodlaender, H.L.: A partial k-arboretum of graphs with bounded treewidth. Theoretical Computer Science 209(1-2), 1–45 (1998)MathSciNetMATHCrossRefGoogle Scholar
- 4.Dell, H., van Melkebeek, D.: Satisfiability allows no nontrivial sparsification unless the polynomial-time hierarchy collapses. In: Symposium on Theory of Computing (STOC), pp. 251–260. ACM (2010)Google Scholar
- 5.Fischer, E., Makowsky, J.A., Ravve, E.V.: Counting truth assignments of formulas of bounded tree-width or clique-width. Discrete Applied Mathematics 156(4), 511–529 (2008)MathSciNetMATHCrossRefGoogle Scholar
- 6.Georgiou, K., Papakonstantinou, P.A.: Complexity and Algorithms for Well-Structured k-SAT Instances. In: Kleine Büning, H., Zhao, X. (eds.) SAT 2008. LNCS, vol. 4996, pp. 105–118. Springer, Heidelberg (2008)CrossRefGoogle Scholar
- 7.Impagliazzo, R., Paturi, R.: Complexity of k-sat. In: Proceedings of Fourteenth Annual IEEE Conference on Computational Complexity, pp. 237–240. IEEE (1999)Google Scholar
- 8.Impagliazzo, R., Paturi, R., Zane, F.: Which problems have strongly exponential complexity? Journal of Computer and System Sciences (JCSS) 63(4), 512–530 (2001); (also FOCS 1998)MathSciNetMATHCrossRefGoogle Scholar
- 9.Kloks, T.: Treewidth: computations and approximations, vol. 842. Springer (1994)Google Scholar
- 10.Lokshtanov, D., Marx, D., Saurabh, S.: Known algorithms on graphs of bounded treewidth are probably optimal. In: 22nd ACM/SIAM Symposium on Discrete Algorithms (SODA 2011), pp. 777–789 (2011)Google Scholar
- 11.Papakonstantinou, P.A.: A note on width-parameterized sat: An exact machine-model characterization. Information Processing Letters (IPL) 110(1), 8–12 (2009)MathSciNetMATHCrossRefGoogle Scholar
- 12.Samer, M., Szeider, S.: A fixed-parameter algorithm for# sat with parameter incidence treewidth. Arxiv preprint cs/0610174 (2006)Google Scholar
- 13.Szeider, S.: On Fixed-Parameter Tractable Parameterizations of SAT. In: Giunchiglia, E., Tacchella, A. (eds.) SAT 2003. LNCS, vol. 2919, pp. 188–202. Springer, Heidelberg (2004)CrossRefGoogle Scholar