Average case complexity of weighted approximation and integration over \(\mathbb R_{+}\) (Q700174)

The authors study weighted approximation and integration of Gaussian stochastic processes \(X\) defined over \(R_+\), whose \(r\)th derivative satisfy a Hölder condition with exponent \(\beta\) in the quadratic mean. It is assumed that that the algorithms use samples of \(X\) at a finite number of points. The average case (information) complexity is studied, i.e. the minimal number of samples that are sufficient to approximate/integrate \(X\) with the expected error not exceeding \(\varepsilon\). Sufficient conditions (in terms of the weight and the parameters \(r\) and \(\beta\)) for the weighted approximations and weighted integration problems are provided to have finite complexity. For approximations, these conditions are necessary as well. Sufficient conditions are also provided for these complexities to be proportional to the complexities of the corresponding problems defined over \([0,1]\), i.e. proportional to \(\epsilon^{-1/\alpha}\), where \(\alpha=r+\beta\) for the approximation and \(\alpha=r+\beta+1/2\) for the integration.