![]()
*This is the first in a series of colloquium talks on the background, meaning and importance of the seven "Million Dollar Problems"
There are more refined estimates of codimensions that take special features of the variety into account and contain Krull's theorem as a particular case. One of these results is the generalized principal ideal theorem of Eisenbud-Evans and Bruns, which gives information on the codimension of the zero locus of sections of vector bundles. It has been used to estimate the codimension of determinantal loci of a matrix in terms of the size and the rank of the matrix. This in turn gives criteria for a vector bundle to split as a direct sum of line bundles. Although these estimates are sharp in general, one should expect much better results for determinantal loci on smoothvarieties.
This was one of the questions that motivated the recent joint work with David Eisenbud and Craig Huneke I will report on. We extend the Eisenbud-Evans-Bruns theorem and establish bounds for the codimension of determinantal loci that take into account not only the size and the rank rof the matrix, but also the embedding codimension of the ambient ring. This indeed leads to improved estimates over regular rings and generalizes work by Faltings, who had considered the case of rby rminors.
Items for inclusion in the Weekly Calendar can be submitted using this submission form, or send complete information to Hilda Britt. Deadline for inclusion in the weekly calendar is noon every Wednesday. Speakers are encouraged to provide abstracts. Please send items for inclusion in "Announcements" and "Conferences" to the webmaster.
Updated October 25, 2000