Projects per year
We propose a randomized first order optimization algorithm GradientProjection Iterative Sketch (GPIS) and an accelerated variant forefficiently solving large scale constrained Least Squares (LS). Weprovide theoretical convergence analysis for both proposed algorithmsand demonstrate our methods' computational efficiency compared toclassical accelerated gradient method, and the state of the artvariance-reduced stochastic gradient methods through numericalexperiments in various large synthetic/real data sets.
|Publication status||Published - Aug 2017|
|Event||34th International Conference on Machine Learning (ICML), 2017 - Sydney, Australia|
Duration: 6 Aug 2017 → 11 Aug 2017
|Conference||34th International Conference on Machine Learning (ICML), 2017|
|Period||6/08/17 → 11/08/17|
FingerprintDive into the research topics of 'Gradient Projection Iterative Sketch for Large Scale Constrained Least-Squares'. Together they form a unique fingerprint.
1/09/16 → 31/08/22
1/01/15 → 31/12/18