Bounding the number of remarkable values via Jouanolou's theorem (Q2017886): Difference between revisions

In [C. R. Acad. Sci., Paris 112, 761--764 (1891; JFM 23.0319.01)], \textit{H. Poincaré} introduced the so called \textit{remarkable values} for indecomposable rational first integrals \(f/g\) of complex polynomial vector fields defined in \(\mathbb{C}^2\). These values correspond to the complex projective values \((\lambda:\mu)\) such that \(\lambda f-\mu g\) is reducible in \(\mathbb{C}^2[x,y]\) or \(\deg(\lambda f-\mu g)=\deg(f/g)=\max(\deg (f),\deg(g)).\) For vector fields of degree \(k\), it is known that there are at most \(k^2+2\) remarkable values. In some previous works, the author diminishes the upper bound of the number of needed algebraic invariant curves to prove the existence of a Darboux first integral for sparse vector fields by using properties of the Newton polygon associated to them and Jouanolou's theorem. In this work, by using similar ideas the maximum number of remarkable values is bounded by a linear upper bound in \(k\) for some sparse vector fields.