The qTSPP Theorem
Manuel Kauers
Last modified: 2010-03-27
Abstract
The qTSPP Conjecture has been a challenging open
problem in the theory of integer partitions for about 30
years. It was recently turned into a theorem by C. Koutschan,
D. Zeilberger, and the speaker. In a sense the proof consisted of
finding a certificate function which asserts a certain
determinant identity. But our certificate is so complicated that
it is not easy to see that it actually does have all the
properties it needs for being a certificate. So we also computed
a certificate for the certificate. This "meta certificate"
reduces the proof of the determinant conjecture to a certain
amount of basic arithmetic.
Our proof of the qTSPP conjecture is noteworthy not only for its
significance in combinatorics. It is noteworthy also for the
enourmous computational effort that was needed to get the
computations done. It illustrates again that computer algebra
algorithms for special functions are not only interesting but
also useful. Specifically, it also shows that---despite involving
expensive operations such as Groebner basis computations in
operator algebras---even calculations of scaringly large size can
be brought to a successful end.
problem in the theory of integer partitions for about 30
years. It was recently turned into a theorem by C. Koutschan,
D. Zeilberger, and the speaker. In a sense the proof consisted of
finding a certificate function which asserts a certain
determinant identity. But our certificate is so complicated that
it is not easy to see that it actually does have all the
properties it needs for being a certificate. So we also computed
a certificate for the certificate. This "meta certificate"
reduces the proof of the determinant conjecture to a certain
amount of basic arithmetic.
Our proof of the qTSPP conjecture is noteworthy not only for its
significance in combinatorics. It is noteworthy also for the
enourmous computational effort that was needed to get the
computations done. It illustrates again that computer algebra
algorithms for special functions are not only interesting but
also useful. Specifically, it also shows that---despite involving
expensive operations such as Groebner basis computations in
operator algebras---even calculations of scaringly large size can
be brought to a successful end.