2
$\begingroup$

Let $R = k[x_1\ldots x_n]$, and $a,b$ are vectors with integer entries, whose all entries of $a$ are non-negative, and say sum of coordinates of $b$ is $0$. Let $I$ be a monomial ideal generated by $x^a, x^{a+b}, \ldots, x^{a + kb}$ whenever $a + kb$ has non-negative integer entries. Is this true that regularity of $I$ is at most two times the degree of $a$? More generally, if we have $c$ pairs of vectors $a_1, b_1, \ldots, a_c, b_c$ and consider monomial ideal generated by the powers of translations of the $a_i$ by $b_i$. Is this true that the regularity of $I$ is at most two times the sum of degrees of $a_i$?

$\endgroup$
2
$\begingroup$

Doesn't a "yes" answer for your first question follow straightforwardly from facts about the lcm lattice?

See e.g. Nevo's "Regularity of edge ideals of $C_4$ free graphs via the topology of the lcm lattice" equation (1) for a clear statement of the relationship of regularity to the lcm lattice. In particular, (excluding the case where I has a single generator)

$\mathrm{reg}\ I \leq \max \deg m,$ where $m$ varies over proper least common multiples of generators of $I$.

In this case, the lcm of $x^{a+ib}, x^{a+jb}$ and $x^{a+\ell b}$ where $i<j<\ell$ is the same as that of $x^{a+ib}$ and $x^{a+\ell b}$, hence the greatest degree of an lcm is that of the lcm of $x^a$ and $x^{a+kb}$. By the zero sum condition, this is at most twice the degree of $x^a$.

There's almost certainly a more direct way to see the fact from Nevo's paper that I used.

A "yes" answer to the latter question should then follow via the Kalai-Meshulam Theorem (see "Intersections of Leray complexes and regularity of monomial ideals" Theorem 1.4; I also discuss this theorem in my own paper "Matchings, coverings and CM regularity").

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.