ID: a software package for low-rank approximation of matrices via interpolative decompositions. This software distribution provides Fortran routines for computing low-rank approximations to matrices, in the forms of interpolative decompositions (IDs) and singular value decompositions (SVDs). The routines use algorithms based on the ID. The ID is also commonly known as the approximation obtained via skeletonization, the approximation obtained via subsampling, and the approximation obtained via subset selection. The ID provides many advantages in many applications, and we suspect that it will become increasingly popular once tools for its computation become more widely available. This software distribution includes some such tools, as well as tools for computing low-rank approximations in the form of SVDs. Section 8 below defines IDs and SVDs, and provides references to detailed discussions of the algorithms used in this software package ...
Keywords for this software
References in zbMATH (referenced in 4 articles )
Showing results 1 to 4 of 4.
- Tygert, Mark: Regression-aware decompositions (2019)
- Pereyra, Victor: Model order reduction with oblique projections for large scale wave propagation (2016)
- Saibaba, Arvind K.: HOID: higher order interpolatory decomposition for tensors based on Tucker representation (2016)
- Gu, M.: Subspace iteration randomization and singular value problems (2015)