Kähler-Einstein metrics on Fano manifolds (Q2341298): Difference between revisions

The author's summary: ``This is an expository paper on Kähler metrics of positive scalar curvature. It is for my Takagi Lectures at RIMS in November of 2013. In this paper, I first discuss the Futaki invariants, the K-stability and its relation to the K-energy. Next, I will outline my work in 2012 [Commun. Pure Appl. Math. 68, No. 7, 1085--1156 (2015; Zbl 1318.14038)] on the existence of Kähler-Einstein metrics on K-stable Fano manifolds. Finally, I will present \textit{S. Paul}'s work [``Stable pairs and coercive estimates for the Mabuchi functional'', Preprint, \url{arXiv:1308.4377}] on stability of pairs with some modifications of mine.'' The author is one of the major experts in compact Kähler-Einstein metrics with positive Einstein constant. This is an exposition of his work on the Yau-Tian-Donaldson conjecture in [the author, Commun. Pure Appl. Math. 68, No. 7, 1085--1156 (2015; Zbl 1318.14038)]. Notice that that paper appeared in the same journal as Professor Yau's famous paper [\textit{S.-T. Yau}, Commun. Pure Appl. Math. (1978; Zbl 0362.53049)] on the Calabi conjectures. One might consult the review of that paper. The problem of finding Kähler-Einstein metrics on Fano manifolds is one of the most difficult problems in Riemannian geometry. From the point of view of the reviewer, it is more difficult than the negative and zero cases. And it is not completely solved so far. The author points (Remark 9) that Donaldson's original definition of K-stability needs to be fixed. For the details see [\textit{C. Li} and \textit{C. Xu}, Ann. Math. (2) 180, No. 1, 197--232 (2014; Zbl 1301.14026)]. This is one of the questions the reviewer asked Donaldson many years ago. According to X. X. Chen, to solve this kind of PDE problem in Riemannian geometry, it is wise to do the cohomogeneity-one case first and then extend it to the general case. This might be the point of view from as early as Calabi. For the case with positive scalar curvature, and the compact case, the problem of finding extremal metrics was completely solved by the reviewer [with \textit{X. Chen}, Asian J. Math. 4, No. 4, 817--830 (2000; Zbl 1003.32003); Int. J. Math. 14, No. 3, 259--287 (2003; Zbl 1048.32014); ``Type I almost-homogeneous manifolds of cohomogeneity one. I'', Pac. J. Appl. Math. 3, No. 1--2, 43--72 (2011); Pac. J. Math. 261, No. 2, 369--388 (2013; Zbl 1288.14032)]. The first step of the blowup analysis was suggested by Professor Paul Yang after the reviewer tried an approach from dynamical systems by discussing with Professor Mather. We actually used a kind of conic Einstein metrics at the beginning (the second step), noticing that the divisors we used in the cohomogeneity-one case are Fano submanifolds. This is consistent with Nadel's results saying that the obstruction to the existence of Kähler-Einstein metrics comes from Fano-like subvarieties. It is a little bit hard for the reviewer to understand that people could use highly negative hyperbolic divisors for the conic Kähler-Einstein metrics. After we finished approaching the smooth Kähler-Einstein metrics by the conic Kähler-Einstein metrics, one could see that one could actually approach Aubin's metrics (with \(t<1\)) by the conic metrics also. The existence of Aubin metrics is determined by a partial Futaki invariant. That is, one approaches smooth Aubin metrics by conic Aubin metrics. The exactly proof goes through. Therefore, if the conic method goes through, the original Aubin method should go through also. To our encouragement, Gábor Székelyhidi did claim that the Aubin's method did go through if the automorphism group is finite. It is going to appear as [\textit{G. Székelyhidi}, ``The partial \(C^0\)-estimate along the continuity method'', Preprint, \url{arXiv:1310.8471}]. While reviewing the present paper, we asked one of our colleagues (in the University of California): ``where is the one with nontrivial holomorphic vector fields?'' We then recently found these two papers: [\textit{V. Datar} and \textit{G. Székelyhidi}, ``Kähler-Einstein metrics along the smooth continuity method'', Preprint, \url{arXiv:1506.07495}] and [\textit{N. Ilten} and \textit{H. Süss}, ``K-stability for Fano manifolds with torus action of complexity one'', Preprint, \url{arXiv:1507.04442}]. As these authors are seemingly trying to avoid mentioning the results in the cohomogeneity-one case and Nadel's result on the Fano-like subvariety obstructions, one might wonder whether these results are consistent with the cohomogeneity-one case and Nadel's result. It is not difficult to see that we shall have a clearer picture in the near future.