English

Computing gaussian \& exponential measures of semi-algebraic sets

Optimization and Control 2017-07-11 v2

Abstract

We provide a numerical scheme to approximate as closely as desired the Gaussian or exponential measure μ(\om)\mu(\om) of (not necessarily compact) basic semi-algebraic sets\omRn\om\subset\R^n. We obtain two monotone (non increasing and non decreasing) sequences of upper and lower bounds (ω_d)(\overline{\omega}\_d), (ω_d)(\underline{\omega}\_d), dNd\in\N, each converging to μ(\om)\mu(\om) as dd\to\infty. For each dd, computing ω_d\overline{\omega}\_d or ω_d\underline{\omega}\_dreduces to solving a semidefinite program whose size increases with dd. Some preliminary (small dimension) computational experiments are encouraging and illustrate thepotential of the method. The method also works for any measure whose moments are known and which satisfies Carleman's condition.

Keywords

Cite

@article{arxiv.1508.06132,
  title  = {Computing gaussian \& exponential measures of semi-algebraic sets},
  author = {Jean-Bernard Lasserre},
  journal= {arXiv preprint arXiv:1508.06132},
  year   = {2017}
}

Comments

To appear in Advances in Applied Mathematics

R2 v1 2026-06-22T10:41:02.321Z