Valuations on convex functions and convex sets and Monge-Ampère operators (Q2322402)

The notion of a valuation on convex bodies is very classical; valuations on a class of functions have been introduced and studied by M. Ludwig and others. The author studies an explicit relation between continuous valuations on convex functions which are invariant under adding arbitrary linear functionals, and translation invariant continuous valuations on convex bodies. More precisely, he constructs a natural linear map \(T\) from the function-space to the space of valuations and proves that \(T\) has dense image and infinite-dimensional kernel. The proof uses the author's irreducibility theorem and properties of the real Monge-Ampère operators due to A. D. Alexandrov and Z. Blocki. For the proof the space of valuations is separated into the direct sum of spaces of valuations of given dimension. Then the main claim that the image of the linear map \(T\) within the \(i\)-dimensional valuations is never void is proved by induction over the dimension. (To do so the author constructs for each \(i\)-dimensional space of valuations members that is an element of im(\(T\))). Finally from the claim the theorem is derived using smoothing methods. The first major result used is a theorem of \textit{S. Alesker} [Geom. Funct. Anal. 11, No. 2, 244--272 (2001; Zbl 0995.52001)] that for any convex function \(f\) one can define a signed measure that connect the Hessian of \(f\) with its Monge-Ampère operator. The second major result of \textit{Z. Blocki} [Proc. Am. Math. Soc. 128, No. 12, 3595--3599 (2000; Zbl 0959.32039)] assures that this measure satisfies the necessary valuation properties. Furthermore the author shows how to use complex, quaternionic, and octonionic Monge-Ampère operators and their Hessian to construct more examples of continuous valuations on convex functions in an analogous way. In general the author defines these operators and extend this result to these spaces. It would be interesting to know, if there is a simple connection of these results to the Monge-Ampère operator in multidimensional real space.