$S_{k,\exp}$ does not prove $\mathrm{NP}=\mathrm{co}-\mathrm{NP}$ uniformly
Zapiski Nauchnykh Seminarov POMI, Computational complexity theory. Part VIII, Tome 304 (2003), pp. 99-120
Voir la notice de l'article provenant de la source Math-Net.Ru
A notion of a uniform sequent calculus proof is given. It is then shown that a strengthening, $S_{k,\exp}$, of the well-studied bounded arithmetic system $S_k$ of Buss does not prove $\mathrm{NP}=\mathrm{co}-\mathrm{NP}$ with a uniform proof. A slightly stronger result that $S_{k,\exp}$ cannot prove $\widehat\Sigma_{1,k'}^b=\widehat\Pi_{1,k'}^b$ uniformly for $2\leq k'\leq k$ is
also established. A variation on the technique used is then applied to show that $S_{k,\exp}$ is unable to prove
Davis–Putnam–Robinson–Matiyasevich theorem. This result is also without any uniformity conditions.
Generalization of both these results to higher levels of the Grzegorczyck Hierarchy are then presented.
@article{ZNSL_2003_304_a5,
author = {Ch. Pollett},
title = {$S_{k,\exp}$ does not prove $\mathrm{NP}=\mathrm{co}-\mathrm{NP}$ uniformly},
journal = {Zapiski Nauchnykh Seminarov POMI},
pages = {99--120},
publisher = {mathdoc},
volume = {304},
year = {2003},
language = {en},
url = {http://geodesic.mathdoc.fr/item/ZNSL_2003_304_a5/}
}
Ch. Pollett. $S_{k,\exp}$ does not prove $\mathrm{NP}=\mathrm{co}-\mathrm{NP}$ uniformly. Zapiski Nauchnykh Seminarov POMI, Computational complexity theory. Part VIII, Tome 304 (2003), pp. 99-120. http://geodesic.mathdoc.fr/item/ZNSL_2003_304_a5/