Homogeneous spaces and equivariant embeddings (Q621541): Difference between revisions

The book under review is concerned with the study of equivariant embeddings of homogeneous spaces: if \(G/H\) is a homogeneous \(G\)-space, an equivariant embedding of \(G/H\) is a \(G\)-variety \(X\) containing \(G/H\) as a dense open orbit. Given a homogeneous space \(G/H\), one would (to the extent possible) like to classify all such embeddings, as well as study their geometric properties. Common examples of such equivariant embeddings are flag varieties, toric varieties, and spherical varieties. This monograph provides a compact survey of these issues, including many results of Brion, Knop, Luna, and Vust, among others, as well as the author's own significant contributions to the field. At the heart of the text is the Luna-Vust theory, which classifies equivariant embeddings of \(G/H\) using `combinatorial' language involving \(B\)-invariant divisors of \(G/H\) (here \(B\) denotes a Borel subgroup of \(G\)) and \(G\)-invariant valuations of the function field of \(G/H\). The author actually presents his own generalization of this theory in which he replaces \(G/H\) by any \(G\)-variety. In general, the difficulty of describing the \(B\)-invariant divisors and \(G\)-invariant valuations makes such classification impossible in practice, but for special cases, this does become tractable. Indeed, if the complexity of the \(G\)-action (defined to be the codimension of a general \(B\)-orbit) is 0 or 1, the above classification becomes genuinely combinatorial. The author devotes considerable time to these cases, in which one may also describe many geometric aspects of a given equivariant embedding. The author wisely assumes that the reader has a background in algebraic geometry and representation theory; this keeps the length of the monograph manageable. The book begins with a basic discussion of algebraic homogeneous spaces in Chapter 1. Chapter 2 deals with two invariants of a \(G\)-variety \(X\): its complexity (defined above) and its rank (the rank of the lattice of weights of \(B\)-eigenfunctions). The main results are general formulae for computing these invariants. Chapter 3 contains the author's presentation of the Luna-Vust theory mentioned above. Particular attention is paid to the complexity 0 and complexity 1 cases. A general description of \(B\)-stable divisors is given, which is again refined for the complexity 0 and 1 cases. In these cases, the author also presents a number of results on intersection theory. Chapter 4 contains a discussion of \(G\)-invariant valuations, important due to their role in the Luna-Vust theory. Finally, Chapter 5 returns to the study of complexity 0 \(G\)-varieties, also known as spherical \(G\)-varieties. A number of characterizations of such varieties are presented, as well as descriptions of some special subclasses, and results on Frobenius splittings. The monograph also includes several appendices which cover a number of topics in algebraic geometry and representation theory which are used elsewhere in the text. The author admirably fills a gap in the current literature on algebraic groups with this book. It should make a welcome addition to any mathematician's library.