Refining logical characterizations of advice complexity classes

Numerical relations in logics are known to characterize, via the finite models of their sentences, polynomial advice nonuniform complexity classes. These are known to coincide with reduction classes of tally sets. Our contributions here are: 1/ a refinement of that characterization that individualiz...

Full description

Bibliographic Details
Authors: Atserias, Albert|||0000-0002-3732-1989, Balcázar Navarro, José Luis|||0000-0003-4248-4528
Format: report
Publication Date:1997
Country:España
Institution:Universitat Politècnica de Catalunya (UPC)
Repository:UPCommons. Portal del coneixement obert de la UPC
Language:English
OAI Identifier:oai:upcommons.upc.edu:2117/83486
Online Access:https://hdl.handle.net/2117/83486
Access Level:Open access
Keyword:Nonuniform complexity classes
Computational complexity
Àrees temàtiques de la UPC::Informàtica::Informàtica teòrica
Description
Summary:Numerical relations in logics are known to characterize, via the finite models of their sentences, polynomial advice nonuniform complexity classes. These are known to coincide with reduction classes of tally sets. Our contributions here are: 1/ a refinement of that characterization that individualizes the reduction class of each tally set, and 2/ characterizing logarithmic advice classes via numerical constants, both in the (rather easy) case of C/log and in the more complex case of Full-C/log; this proof requires to extend to classes below P the technical characterizations known for the class Full-P/log.