Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review
The power of negative reasoning. / de Rezende, Susanna F.; Lauria, Massimo; Nordström, Jakob; Sokolov, Dmitry.
36th Computational Complexity Conference, CCC 2021. ed. / Valentine Kabanets. Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing, 2021. 40 (Leibniz International Proceedings in Informatics, LIPIcs; Vol. 200).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review
}
TY - GEN
T1 - The power of negative reasoning
AU - de Rezende, Susanna F.
AU - Lauria, Massimo
AU - Nordström, Jakob
AU - Sokolov, Dmitry
N1 - Publisher Copyright: © Susanna F. de Rezende, Massimo Lauria, Jakob Nordström, and Dmitry Sokolov;
PY - 2021/7/1
Y1 - 2021/7/1
N2 - Semialgebraic proof systems have been studied extensively in proof complexity since the late 1990s to understand the power of Gröbner basis computations, linear and semidefinite programming hierarchies, and other methods. Such proof systems are defined alternately with only the original variables of the problem and with special formal variables for positive and negative literals, but there seems to have been no study how these different definitions affect the power of the proof systems. We show for Nullstellensatz, polynomial calculus, Sherali-Adams, and sums-of-squares that adding formal variables for negative literals makes the proof systems exponentially stronger, with respect to the number of terms in the proofs. These separations are witnessed by CNF formulas that are easy for resolution, which establishes that polynomial calculus, Sherali-Adams, and sums-of-squares cannot efficiently simulate resolution without having access to variables for negative literals.
AB - Semialgebraic proof systems have been studied extensively in proof complexity since the late 1990s to understand the power of Gröbner basis computations, linear and semidefinite programming hierarchies, and other methods. Such proof systems are defined alternately with only the original variables of the problem and with special formal variables for positive and negative literals, but there seems to have been no study how these different definitions affect the power of the proof systems. We show for Nullstellensatz, polynomial calculus, Sherali-Adams, and sums-of-squares that adding formal variables for negative literals makes the proof systems exponentially stronger, with respect to the number of terms in the proofs. These separations are witnessed by CNF formulas that are easy for resolution, which establishes that polynomial calculus, Sherali-Adams, and sums-of-squares cannot efficiently simulate resolution without having access to variables for negative literals.
KW - Nullstellensatz
KW - Polynomial calculus
KW - Proof complexity
KW - Sherali-Adams
KW - Sums-of-squares
UR - http://www.scopus.com/inward/record.url?scp=85115341992&partnerID=8YFLogxK
UR - https://www.mendeley.com/catalogue/f562c232-fe39-32d0-9bff-03d9eb0f9b32/
U2 - 10.4230/LIPIcs.CCC.2021.40
DO - 10.4230/LIPIcs.CCC.2021.40
M3 - Conference contribution
AN - SCOPUS:85115341992
SN - 9783959771931
T3 - Leibniz International Proceedings in Informatics, LIPIcs
BT - 36th Computational Complexity Conference, CCC 2021
A2 - Kabanets, Valentine
PB - Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing
T2 - 36th Computational Complexity Conference, CCC 2021
Y2 - 20 July 2021 through 23 July 2021
ER -
ID: 87790148