Quantum reduction and representation theory of superconformal algebras (Q596613)

In an earlier paper with \textit{E. Frenkel} [Commun. Math. Phys. 147, 295--328 (1992; Zbl 0768.17008)] the authors studied certain \(W\)-algebras obtained by (quantum) reduction from representations of affine Lie algebras. The construction is homological where the \(W\)-algebra is obtained as the 0-th group. In the same papers several conjectures are stated relating `fusion rings' of \(W\)-algebras with the fusion rings of WZW models. In the meantime the rapid development of vertex algebra theory, in particular, led to a different interpretation of \(W\)-algebras. Instead of thinking in terms of generators and relations, \(W\)-algebras should be viewed as vertex operator (super)algebras and \(W\)-algebra modules as vertex (super)algebra modules. This setup has a great advantage. In particular one can then make sense of ``fusion rings'', etc. In the paper under review the authors recall their construction from the mentioned paper but now in the setup of vertex algebras. In several cases they are able to give an explicit description of certain \(W\)-algebras in terms of familiar (and well-studied) vertex (super)algebras, such as \(N=2\) superconformal vertex algebras, Neveu-Schwarz vertex algebras, etc. Their main result is Theorem 4.1 which, in a quite general setup, gives a nice description of \(W\)-algebras in terms of certain PBW bases. They also prove that higher homology groups are trivial. These results are very important and can be used for further studies of \(W\)-algebras. The paper also provides numerous examples. In the corrigendum the authors correct the proof of Theorem 4.1.