We propose a stochastic variance reduced optimization algorithm for solving a class of largescale nonconvex optimization problems with cardinality constraints, and provide sufficient conditions under which the proposed algorithm enjoys strong linear convergence guarantees and optimal estimation accuracy in high dimensions. Numerical experiments demonstrate the efficiency of our method in terms of both parameter estimation and computational performance.
|Original language||English (US)|
|Title of host publication||33rd International Conference on Machine Learning, ICML 2016|
|Editors||Maria Florina Balcan, Kilian Q. Weinberger|
|Publisher||International Machine Learning Society (IMLS)|
|Number of pages||13|
|State||Published - 2016|
|Event||33rd International Conference on Machine Learning, ICML 2016 - New York City, United States|
Duration: Jun 19 2016 → Jun 24 2016
|Name||33rd International Conference on Machine Learning, ICML 2016|
|Other||33rd International Conference on Machine Learning, ICML 2016|
|City||New York City|
|Period||6/19/16 → 6/24/16|
Bibliographical noteFunding Information:
This research is supported by NSF CCF-1217751; NSF AST-1247885; DARPA Young Faculty Award N66001-14-1-4047; NSF DMS-1454377-CAREER; NSF IIS-1546482-BIGDATA; NIH R01MH102339; NSF IIS-1408910; NSFIIS-1332109; NIH R01GM083084.