Approximations via (lower-set) Least-Squares Adaptive Chaos Expansions (ALSACE)
Development/maintenance:
Armin Herbert Galetzka ([email protected])
Dimitrios Loukrezis ([email protected])
ALSACE is a Python software for multivariate approximation based on polynomial chaos expansions (PCEs). The PCE coefficients are computed using least squares (LS) regression. The PCE polynomial basis as well as the experimental design used in the LS problem are expanded adaptively by exploiting parameter anisotropies and LS stability estimates. The software has been developed as part of our work at the Institute for Accelerator Science and Electromagnetic Fields (TEMF) of the Technische Universität Darmstadt.
The ALSACE software has been employed in the following works:
We kindly ask you to cite at least one of those works, in case you use ALSACE for your own research.
The present software and the related examples rely partially on the OpenTURNS C++/Python library.
- http://www.openturns.org/
- Open TURNS: An industrial software for uncertainty quantification in simulation, https://arxiv.org/abs/1501.05242
Option "AI": Sequential Experimental Designs based on A/I-optimality criteria
Option "K": Sequential Experimental Designs based on K-optimality criteria
Option "E": Sequential Experimental Designs based on E-optimality criteria