ABSTRACT
ABSTRACT: Polynomial chaos expansion have received much attention in the past ten years in the domain of computational stochastic mechanics. One keypoint in this context is the problem of the curse of dimensionality. In order to bypass this difficulty, the present paper proposes the use of sparse polynomial chaos basis built from a prior choice of monomials by the so-called hyperbolic q-norms and a selection algorithm known as Least Angle regression. The approach is illustrated by solving a structural reliability problem of a nuclear reactor vessel submitted to a pressurized thermal shock.
