The multidimensional truncated moment problem: the moment cone (Q2669362)

Given a finite set \(A=\{a_1,\ldots,a_m\}\) of measurable functions on a measurable space \((\mathcal X,\mathcal U)\), the paper deals with the analysis of some properties of the corresponding truncated moment cone \(\mathcal S_A\), i.e., the set of all sequences \((s_1,\ldots,s_m)\) obtained as integrals \(s_i=\int_{\mathcal X} a_i d\mu\) for some positive measure \(\mu\) on \((\mathcal X,\mathcal U)\), called \textit{moment sequences}. The main two properties adressed in the paper are the following: \begin{itemize} \item[1.] Understand the facial structure of \(\mathcal S_A\) regarding exposed faces and the dimension of faces. \item[2.] Bound the Caratheodory number of \(\mathcal S_A\), being the smallest number of atoms in some representing measure \(\mu\) of any sequence from \(\mathcal S_A\). \end{itemize} Except the two problems above the other results of the paper are related to the following three problems: \begin{itemize} \item[(a)] For \(\mathcal X\subseteq \mathbb{R}^n\) and \(a_i\in C^1(\mathcal X,\mathbb R)\) understand the differential structure of the moment map and the smoothness properties of the moment sequences. \item[(b)] The maximal mass problem. \item[(c)] Applications to related problems. \end{itemize} The following paragraph describes more precisely the organization of the paper. The introductory section gives a brief history of \textit{the Moment Problem (MP)} with a lot of references to the existing literature and a description of various applications. Sections 2 and 3 are preparatory containing necessarry definitions and preliminary results. Section 4 is then devoted to the problem 3. Section 5 gives a new proof and a detailed history of an important result, the authors call \textit{Richter's theorem}, which states that every truncated moment sequence admits a finitely atomic representing measure. Problem 1 is the contents of Section 6, which is the richest section of the paper, containing a thorough analysis of the faces of \(\mathcal S_A\). Various numbers are defined related to the dimension of the faces and statistically analysed for different values of the parameters. Section 7 is devoted to the problem 2, deriving new lower and upper bounds on the Caratheodory number. Sections 8-10 are shorter and devoted to problems 3-5. Precisely, first the second question of the problem 3 is studied mainly based on concrete examples, second the applications of the main results of the paper in two related questions (understanding sum of squares and tensor decompositions) are given and last problem 4 in investigated. Finally, Section 11 lists many open problems related to the results of the paper. The paper is well written in a self-contained manner, being accessible also to non-experts in the field.