Abstract
We examine here the existence of approximations in NC of P-complete problems. We show that many P-complete problems (such as UNIT, PATH, circuit value etc.) cannot have an approximating solution in NC for any value of the absolute performance ratio R of the approximation, unless P=NC. On the other hand, we exhibit of a purely combinatorial problem (the High Connectivity subgraph problem) whose behaviour with respect to fast parallel approximations is of a threshold type. This dichotomy in the behaviour of approximations of P-complete problems is for the first time revealed and we show how the tools of log-space reductions can be used to make inferences about the best possible performance of approximations of problems that are hard to parallelize.
This research was done during the visit of the first author to Patras University, and is supported partially by a Spanish Research Scholarship and by the Ministry of Industry, Energy and Technology of Greece.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
"Complete Problems for Deterministic Polynomial Time" by N. Jones and W. Laaser, Theor. Computer Science 3 (1977), 105–117.
"Optimization, Approximation and Complexity classes", by C.H. Papadimitriou and M. Yannakakis, STOC 1988.
"Approximating P-Complete Problems", by R. Anderson and E. Mayr, TR Stanford Univ., 1986.
"Parallel Approximation Algorithms for one-Dimensional Bin Packing", by Warmuth, TR Univ. of California at Santa Cruz, 1986.
"Probabilistic log-space reductions and problems probabilistically hard for P", by L. Kirousis and P. Spirakis, SWAT 88.
"On the Structure of Combinatorial Problems and Structure Preserving Reductions", by Ansiello, D'Atri and Protasi, Proc. 4th ICALP 45–57, 1977.
"Structure Preserving Reductions among Convex Optimization Problems", by Ansiello, D'Atri and Protasi, J. Comp. Sys. Sc. 21, 136–153, 1980.
"Toward a Unified Approach for the Classification of NP-Complete Optimization Problems", by Ansiello, Marchetti-Spaccamela and Protasi, 12, 83–96, 1980.
"Computers and Intractability — A Guide to the Theory of NP-Completeness", by M. Garey and D. Johnson, Freeman, 1979.
"Non Deterministic Polynomial Optimization Problems and their Approximation", by A. Paz and S. Moran, Theor. Computer Science, 15, 251–277, 1981.
"k-blocks and ultrablocks in graphs", by D. Matula, J. of Combinatorial Theory, 24, 1978, pp. 1–13.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1989 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Serna, M., Spirakis, P. (1989). The approximability of problems complete for P. In: Djidjev, H. (eds) Optimal Algorithms. OA 1989. Lecture Notes in Computer Science, vol 401. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-51859-2_16
Download citation
DOI: https://doi.org/10.1007/3-540-51859-2_16
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-51859-4
Online ISBN: 978-3-540-46831-8
eBook Packages: Springer Book Archive