Formality of DG algebras (after Kaledin) (Q2269635)

This paper is devoted to provide the necessary foundations and proofs of the main results by \textit{D. Kaledin} [Mosc. Math. J. 7, No.~4, 643--652 (2007; Zbl 1163.18006)]. It is suggested by the many misprints and inaccuracies in the quoted paper and also by unsatisfactory definitions and approaches in it. If \(K\) is a field of characteristic zero and \(A\) is a \(DG\) algebra over \(K\), Kaledin defines a cohomology class which is the obstruction to the formality of \(A\). This class is actually an element of the second Hochschild cohomology group of a \(DG\)-algebra \(\tilde{A}\) which is closely related to \(A\). In the paper under review the author found more convenient to work with \(A(\infty)\) algebras rather than with \(DG\) algebras and for them he considers Hochschild cohomology with compact supports. These cohomology groups contain Kaledin's cohomology classes, which are obstructions to formality, and have a good functorial behaviour that allows to prove a faithfully flat base change result for formality.