\(K\)-theory and derived equivalences (Q1884508): Difference between revisions

There are three main results in this paper. They all conclude from equivalence of derived categories to equivalence of various \(K\)-theories of the original categories. Theorem A says that if the derived categories, \(\mathcal{D}_R\) and \(\mathcal{D}_S\) of two rings \(R\) and \(S\) are equivalent as triangulated categories, then the algebraic \(K\)-groups \(K_*(R)\) and \(K_*(S)\) are isomorphic. Theorem B is a generalization: A perfect complex in \(\mathcal{D}_R\) is a complex which is isomorphic to a bounded complex of finitely generated projectives, and \(\mathcal{D}_c(R)\) is the full subcategory of \(\mathcal{D}_R\) of perfect complexes. When \(\mathcal{D}_c(R)\) and \(\mathcal{D}_c(S)\) are equivalent as triangulated categories, not only are the \(K\)-groups isomorphic, but there is a weak equivalence of the \(K\)-theory spaces \(K(R)\) and \(K(S)\). When \(R\) and \(S\) are Noetherian, there is a similar theorem for \(G(R)\) and \(G(S)\), where \(G(R)\) is the \(K\)-theory of the category of finitely generated \(R\)-modules. The proofs are via model categories and one of the main points is, that when two rings \(R\) and \(S\) are derived equivalent (they satisfy the condition in theorem A), then their model categories of chain complexes are equivalent as model categories. A further generalization is made by replacing \(R\)-modules by a co-complete abelian category with a set of small projective strong generators.