Resolving subcategories closed under certain operations and a conjecture of Dao and Takahashi (Q1980016)

Dao and Takahashi introduced in the last decade the notion of radius of a subcategory of the module category of a commutative Noetherian local ring. Roughly speaking, it is defined as the least number of extensions necessary to build all the modules in the subcategory out of a single module. This is motivated by the notion of Rouquier's dimension of triangulated categories. A natural focus when studying radius is resolving subcategories (i.e. categories which contain the projective modules and are closed under direct summands, extensions and syzygies). When the ring is Cohen-Macaulay, the category of maximal Cohen-Macaulay modules is an example of such a category. Dao and Takahashi proved for complete intersections that if a resolving subcategory has finite radius, then it has to be contained in the category of maximal Cohen-Macaulay modules. They conjectured that this is the case for all Cohen-Macaulay rings. More precisely, they conjectured that if we take a subcategory of the module category of a Cohen-Macaulay local ring and if we assume that this subcategory has finite radius, contains the regular module and is closed under direct summands, extensions and syzygies, then this subcategory can not contain any objects which are not maximal Cohen-Macaulay modules. The first main contribution of the work of Sadeghi and Takahashi is to prove this conjecture for the following cases: 1) When the subcategory is also closed under duals, 2) When the ring has minimal multiplicity and infinite residue field, 3) When the ring is Gorenstein with codimension at most 4 and each module in the subcategory has bounded Betti numbers. 4) When the ring is Gorenstein with multiplicity (with respect to the maximal ideal) is at most 11 and each module in the subcategory has bounded Betti numbers. Over Gorenstein rings, the category of maximal Cohen-Macaulay modules is closed under cosyzgies. Motivated by this, in the last section, the authors consider resolving subcategories of module categories which are closed under cosyzygies. This is done in the generality of local rings (without the Cohen-Macaulay assumption). The main result of this second part of the paper is that in this situation, there are two possibilities: either the given resolving subcategory contains the residue field or it is contained in the category of totally reflexive modules. A consequence of this result is that the conjecture of Dao and Takahashi holds when our local ring is Gorenstein of positive Krull dimension and the given resolving subcategory of finite radius is closed under cosyzygies.