Lower bounds for Betti numbers of monomial ideals (Q1643553): Difference between revisions

Let \(I\) be a monomial ideal of height \(c\) in a polynomial ring \(S\) over a field \(k\), the Betti number \(\beta_i(S/I)\)) denotes the rank of the \(i\)-th free module appearing in a minimal \(S\)-free resolution of \(S/I\). In this paper under review, the authors prove the following main result: Theorem 1.1. Let \(I\) be a monomial ideal of height \(c\) in a polynomial ring \(S\). If I is not a complete intersection, then \(\Sigma\beta_i(S/I)\geq{2^c+2^{c-1}}\). Furthermore, equality holds if and only if the Betti numbers are \(\{1,3,2\}, \{1, 5, 5, 1\},\) or a extension thereof by tensoring with a Koszul complex. By this when equality holds the generating function for \(\beta_i(S/I)\) is either \((1+3t+2t^2)(1+t)^{c-2}\), or \((1+5t+5t^2+t^3)(1+t)^{c-3}.\) Also, the authors ask the following question: Question 1.2. If I is a homogeneous ideal of height \(c\) in a polynomial ring \(S\) that is not a \(CI\), is \(\beta(S/I)\geq{2^c+2^{c-1}}\)?