On the relation between Gröbner and Pommaret bases (Q1264309): Difference between revisions

A Pommaret basis is a special kind of a, in general, non-reduced Gröbner basis. The interest in this notion comes from the fact that there is an algorithm (Zharkov-Blinkov) which computes Pommaret bases and sometimes performs better than Buchberger's algorithm. Let \(k\) be a field, and let \(I\) be an ideal in the polynomial ring \(k[x_0,\dots,x_n]\), with a given term order. The main result of the paper is that the reduced Pommaret basis of \(I\) and the reduced Gröbner basis of \(I\) coincide if and only if the initial ideal of \(I\) is stable. Recall that a monomial ideal \(J\subset k[x_0,\dots,x_n]\) is stable if for every monomial \(m\in J\), if \(x_j\) is the smallest variable dividing \(m\), then \(x_tm/x_j\) is in \(I\) for every \(x_t> x_j\). If \(\text{char}(k)=0\), the initial ideals of homogeneous ideals in generic coordinates are stable; it follows in particular that such ideals have finite Pommaret basis, which is not the case for every ideal.