1: \begin{abstract}

2:   We investigate the problem of representation of moment sequences by

3:   measures in Polynomial Optimization Problems, consisting in finding

4:   the infimum $f^{*}$ of a real polynomial $f$ on a real semialgebraic set

5:   $S$ defined by a quadratic module $\cQ$. We analyse the

6:   exactness of Moment Matrix (MoM) relaxations,

7:   dual to the Sum of Squares (SoS) relaxations, which are hierarchies

8:   of convex cones introduced by Lasserre to approximate measures and positive polynomials.

9:   We show that the MoM relaxation coincides with the dual of

10:   the SoS relaxation extended with the real radical of the support of

11:   the associated quadratic module $\cQ$.  We prove that the

12:   vanishing ideal of the semialgebraic set $S$ is

13:   generated by the kernel of the Hankel operator associated to a

14:   generic element of the truncated moment cone for a sufficiently high

15:   order of the MoM relaxation.  When the quadratic module $\cQ$ is

16:   Archimedean, we show the convergence, in Hausdorff distance, of the

17:   convex sets of the MoM relaxations to the convex set of probability

18:   measures supported on $S$ truncated in a given degree.  We prove the

19:   exactness of MoM relaxation when $S$ is

20:   finite and when regularity

21:   conditions, known as Boundary Hessian Conditions, hold on the minimizers.

22:   This implies that MoM exactness holds generically. When the set of

23:   minimizers is finite, we describe a MoM relaxation which involves

24:   $f^{*}$, show its MoM exactness and propose a practical algorithm to

25:   achieve MoM exactness. We prove that if the real variety of polar

26:   points is finite then the MoM relaxation extended with the polar

27:   constraints is exact. Effective numerical computations illustrate

28:   this MoM exactness property.

29: \end{abstract}

30: