An idempotent approach to truncated moment problems (Q2249216): Difference between revisions

Roughly speaking, a truncated moment problem consists in the following: given a finite multi-sequence of real numbers \(\gamma = (\gamma_\alpha)_{|\alpha|\leq 2m}\), where \(\gamma_0 > 0\), the \(\alpha\)'s are multi-indices of a fixed length \(n\geq 1\) and \(m\geq 0\) is an integer, one looks for a positive measure \(\mu\) on \(\mathbb{R}^n\) (usually called a representing measure for \(\gamma\), which may always be assumed to be atomic) such that \(\gamma_\alpha =\int t^\alpha d\mu\) for all monomials \(t^\alpha\) with \(|\alpha| \leq 2m\). In previous literature by R. E. Curto and L. A. Fialkow, the approach to truncated moment problems is based on an associated moment matrix, whose positivity and flatness lead to the existence and uniqueness of the solutions. The use of the Riesz functional to solve various moment problems and related topics appears in several works by M. Laurent and B. Mourrain, H. Moeller, M. Putinar and the author. In the paper under review, a new approach to truncated moment problems is given through a concept of idempotent elements with respect to associated unital square positive Riesz functionals. The existence of representing measures for such functionals is characterized via some intrinsic conditions.