Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4661830 | Annals of Pure and Applied Logic | 2015 | 17 Pages |
Abstract
We prove that TNC1, the true universal first-order theory in the language containing names for all uniform NC1 algorithms, cannot prove that for sufficiently large n, SAT is not computable by circuits of size n4kc where kâ¥1,câ¥2 unless each function fâSIZE(nk) can be approximated by formulas {Fn}n=1â of subexponential size 2O(n1/c) with subexponential advantage: Pxâ{0,1}n[Fn(x)=f(x)]â¥1/2+1/2O(n1/c). Unconditionally, V0 cannot prove that for sufficiently large n, SAT does not have circuits of size nlogâ¡n. The proof is based on an interpretation of KrajÃÄek's proof (KrajÃÄek, 2011 [15]) that certain NW-generators are hard for TPV, the true universal theory in the language containing names for all p-time algorithms.
Related Topics
Physical Sciences and Engineering
Mathematics
Logic
Authors
Ján Pich,