Structural properties of bounded relations with an application to NP optimization problems

We introduce the notion bounded relation which comprises most resource bounded reducibilities which can be found in the literature, including non-uniform bounded reducibilities such as ⩽TP/poly. We state conditions on bounded relations which are again satisfied for most bounded reducibilities and which imply that every countable partial ordering can be embedded into every proper interval of the recursive degrees. As corollaries, we obtain that every countable partial ordering can be embedded into every proper interval of (REC,⩽TP/poly, as well as into every proper interval between either two maximization or two minimization problems in the structures NPO,⩽E and NPO,⩽L. For the two latter structures, we show further that the result about embeddings of partial orderings extends to embeddings of arbitrary countable distributive lattices where in addition the least or the greatest element of the lattice can be preserved. Among other corollaries, we obtain that for both structures every non-trivial NP optimization problem bounds a minimal pair. In connection with embeddings into NPO we introduce a representation of maximization or minimization problems by polynomial time computable functions. We consider decidability issues w.r.t. this representation and show for example that there is no effective procedure which decides for a (subrecursive) index of a polynomial time computable function whether the corresponding maximization problem is approximable within a constant factor.

[1]  Klaus Ambos-Spies Sublattices of the Polynomial Time Degrees , 1985, Inf. Control..

[2]  Luca Trevisan,et al.  On Approximation Scheme Preserving Reducibility and Its Applications , 2000, Theory Comput. Syst..

[3]  Luca Trevisan,et al.  On Approximation Scheme Preserving Reducability and Its Applications , 1994, FSTTCS.

[4]  David P. Williamson,et al.  A complete classification of the approximability of maximization problems derived from Boolean constraint satisfaction , 1997, STOC '97.

[5]  Richard E. Ladner,et al.  On the Structure of Polynomial Time Reducibility , 1975, JACM.

[6]  Luca Trevisan,et al.  Structure in Approximation Classes , 1999, Electron. Colloquium Comput. Complex..

[7]  Mihalis Yannakakis,et al.  Optimization, approximation, and complexity classes , 1991, STOC '88.

[8]  Alessandro Panconesi,et al.  Completeness in Approximation Classes , 1989, Inf. Comput..

[9]  Diana Schmidt The Recursion-Theoretic Structure of Complexity Classes , 1985, Theor. Comput. Sci..

[10]  Richard J. Lipton,et al.  On the Structure of Sets in NP and Other Complexity Classes , 1981, Theor. Comput. Sci..

[11]  Wolfgang Merkle A generalized account of resource bounded reducibilities , 1997 .

[12]  José L. Balcázar,et al.  Structural Complexity II , 2012, EATCS.

[13]  Uwe Schöning A Uniform Approach to Obtain Diagonal Sets in Complexity Classes , 1982, Theor. Comput. Sci..

[14]  Kurt Mehlhorn,et al.  Polynomial and abstract subrecursive classes , 1974, STOC '74.

[15]  Wolfgang Merkle Lattice Embeddings for Abstract Bounded Reducibilities , 2002, SIAM J. Comput..

[16]  Uwe Schöning Minimal pairs for P , 1984, Theor. Comput. Sci..

[17]  Rajeev Motwani,et al.  On syntactic versus computational views of approximability , 1994, Proceedings 35th Annual Symposium on Foundations of Computer Science.