Finitistic dimension conjectures via Gorenstein projective dimension (Q2247750)

One of the tools to study a category of finitely generated modules is to look at their homological invariants. Specifically, the projective dimension of a module measures how far it is of being projective. Global measures using the projective dimensions are, for instance, the global dimension of the corresponding algebra and its finitistic dimension. This later dimension is defined to be the maximal of the projective dimension of all modules of finite projective dimensions. It is an open conjecture to decide if the finitistic dimension is finite or not and this has been the source of many investigations. The paper under review brings a very nice introduction where the author discuss many aspects of this conjecture, including also the corresponding invariants for the general case of (not necessarily finitely generated) modules. Returning to the finitely generated case, Auslander and Reiten have shown that if the subcategory \({\mathcal P}_\infty^{\mathrm{fin}}\) given by the modules of finite projective dimension is contravariantly finite, then the finitistic dimension is finite. This is not the end of the conjecture because there are examples where such subcategory is not contravariantly finite. Motivated by this result, the authors study the subcategory \({\mathcal G} {\mathcal P}_\infty^{\mathrm{fin}}\) of the finitely generated modules of finite Gorenstein projective dimensions. They show that if \({\mathcal G} {\mathcal P}_\infty^{\mathrm{fin}}\) is contravariantly finite so is \({\mathcal P}_\infty^{\mathrm{fin}}\). For that, they use recent results and techniques involving cotorsion pairs and tilting theory.