1: \begin{abstract}

2: %In this work we study convergence properties of sparse polynomial approximations for a class of affine parametric saddle point problems that cover, e.g., Stokes equations, mixed formulation of diffusion equations, time-harmonic Maxwell system equations, etc. We consider high or infinite dimensional parametrization of the random input data that affinely depend on a series of both globally and locally supported basis functions.

3: %With suitable sparsity assumptions on the parametrization, we obtain the algebraic convergence rates $O(N^{-r})$ for a sparse Taylor approximation as well as for sparse polynomial interpolation and integration of the parametric solutions. We prove that $r$ only depends on a sparsity parameter in the parametrization of the random input, and in particular does not depend on the number of active parameter dimensions or the number of polynomial terms $N$. These results imply that the sparse polynomial approximations can effectively break the curse of dimensionality, which set the theoretical benchmark for the development and application of practical algorithms, such as adaptive, least-squares, and compressive sensing constructions, for the solution of high or infinite dimensional parametric saddle point problems.  

4: %

5: %%contributions

6: %%

7: %%1. relax the condition for parametrization 

8: %%

9: %%2. generalization to parametric linear PDEs 

10: %%

11: %%3. propose a-priori construction method

12: %%

13: %%4. comparison of different collocation points 

14: %%

15: %%5. primal-dual approach to increase convergence rate 

16: %%

17: %%6. test for problems in very high dimensions ($10^4$)

18: %

19: %\end{abstract}

20: