CSSI_Poster_zhenming_Liu.pdf (3.17 MB)
A high performance suite of SVD related solvers for machine learning
Version 2 2020-02-05, 05:42
Version 1 2020-02-04, 20:02
poster
posted on 2020-02-05, 05:42 authored by Zhenming Liu, Andreas StathopoulosAndreas StathopoulosWe present our recent research progress on joint-optimization for ML algorithm and SVD solver. Our major discovery is that the "stopping criteria" of a SVD algorithm can be directly optimized for downstream ML applications. We will present a few examples, in which when we change the stopping criteria of the "inner SVD algorithm", we see significant performance gain (in running time) in the "outer" ML algorithm.
Funding
Elements: Software: NSCI: A high performance suite of SVD related solvers for machine learning
Directorate for Computer & Information Science & Engineering
Find out more...