RUS  ENG
Full version
JOURNALS // Sibirskie Èlektronnye Matematicheskie Izvestiya [Siberian Electronic Mathematical Reports] // Archive

Sib. Èlektron. Mat. Izv., 2018 Volume 15, Pages 987–995 (Mi semr973)

This article is cited in 8 papers

Mathematical logic, algebra and number theory

On the complexity of formulas in semantic programming

S. Ospichevab, D. Ponomarevabc

a Sobolev Institute of Mathematics, pr. Koptyuga, 4, 630090, Novosibirsk, Russia
b Novosibirsk State University, Pirogova, 2, 630090, Novosibirsk, Russia
c A.P. Ershov Institute of Informatics Systems, pr. Lavrentyeva, 6, 630090, Novosibirsk, Russia

Abstract: We consider the complexity of $\Delta_0$ formulas augmented with conditional terms. We show that for formulas having $n$ bounded quantifiers, for a fixed $n$, deciding the truth in a list superstructure with polynomial computable basic operations is of polynomial complexity. When the quantifier prefix has $n$ alternations of quantifiers, the truth problem is complete for the $n$-th level of the polynomial-time hierarchy. Under no restrictions on the quantifier prefix the truth problem is PSPACE-complete. Thus, the complexity results indicate the analogy between the truth problem for $\Delta_0$ formulas with conditional terms and the truth problem for quantified boolean formulas.

Keywords: semantic programming, list structures, polynomial time/space complexity, $\Delta_0$-formulas.

UDC: 510.5

MSC: 03D15, 68Q15

Received June 28, 2018, published September 10, 2018

Language: English

DOI: 10.17377/semi.2018.15.083



Bibliographic databases:


© Steklov Math. Inst. of RAS, 2024