Active subspace methods in theory and practice: applications to kriging surfaces
This work provides a practical framework for dimension reduction in high-dimensional engineering models, particularly when parameter variations are not aligned with coordinate axes.
The paper introduces active subspace methods for dimension reduction in multivariate functions, demonstrating their application to kriging response surfaces. Applied to an elliptic PDE model with 100 parameters, the method outperforms local sensitivity analysis by exploiting gradient information to identify low-dimensional subspaces.
Many multivariate functions in engineering models vary primarily along a few directions in the space of input parameters. When these directions correspond to coordinate directions, one may apply global sensitivity measures to determine the most influential parameters. However, these methods perform poorly when the directions of variability are not aligned with the natural coordinates of the input space. We present a method to first detect the directions of the strongest variability using evaluations of the gradient and subsequently exploit these directions to construct a response surface on a low-dimensional subspace---i.e., the active subspace---of the inputs. We develop a theoretical framework with error bounds, and we link the theoretical quantities to the parameters of a kriging response surface on the active subspace. We apply the method to an elliptic PDE model with coefficients parameterized by 100 Gaussian random variables and compare it with a local sensitivity analysis method for dimension reduction.