Random fields and the enumerative geometry of lines on real and complex hypersurfaces (Q2314808)

The authors prove that the average number of lines on a random cubic surface in real projective \(3\)-space is \(6\sqrt{2}-3\). More generally, they give a formula for the average number of lines on a random hypersurface of degree \(2n-3\) in real projective \(n\)-space, involving the determinant of a special random matrix. Their techniques can also be applied to hypersurfaces in complex projective space. They express the number of lines on a generic hypersurface of degree \(2n-3\) in complex projective \(n\)-space in terms of the determinant of another special random matrix. In particular, they recover that a generic cubic surface in complex projective \(3\)-space has \(27\) lines. Their results also imply that these real and complex counts satisfy the ``square root law'', i.e., that the average number of real lines on a random real hypersurface is roughly the square root of the generic number of complex lines. In addition, the methods in the article at hand provide an alternative proof for the following statement which was previously proven in [\textit{C. Okonek} and \textit{A. Teleman}, ``Intrinsic signs and lower bounds in real algebraic geometry'', J. Reine Angew. Math. 688, 219--241 (2014; Zbl 1344.14034)] and [\textit{S. Finashin} and \textit{V. Kharlamov}, ``Abundance of real lines on real projective hypersurfaces'', Int. Math. Res. Not. 2013, No. 16, 3639--3646 (2013; Zbl 1312.14132)]: the number of lines, counted with their intrinsic signs, on a generic hypersurface of degree \(2n-3\) in real projective \(n\)-space is \((2n-3)!!\). This number is a lower bound on the average number of real lines on a random real hypersurface. The probabilistic framework introduced by the authors cannot only be applied to the real and complex enumerative geometry of lines on hypersurfaces but is more general. It advances the field of random real algebraic geometry which was pioneered by the Shub-Smale program of studying the real zeros of random polynomial systems, i.e., the real points of intersection of random hypersurfaces in projective space. The results in the paper at hand extend this study to more general ambient spaces than projective space: they are the first results of this type on Grassmannians.