File Download
There are no files associated with this item.
Supplementary
-
Citations:
- Scopus: 0
- Appears in Collections:
Conference Paper: Variance reduction via accelerated dual averaging for finite-sum optimization
Title | Variance reduction via accelerated dual averaging for finite-sum optimization |
---|---|
Authors | |
Issue Date | 2020 |
Citation | Advances in Neural Information Processing Systems, 2020, v. 2020-December How to Cite? |
Abstract | In this paper, we introduce a simplified and unified method for finite-sum convex optimization, named Variance Reduction via Accelerated Dual Averaging (VRADA). In both general convex and strongly convex settings, VRADA can attain an O(n1)-accurate solution in O(n log log n) number of stochastic gradient evaluations which improves the best known result O(n log n), where n is the number of samples. Meanwhile, VRADA matches the lower bound of the general convex setting up to a log log n factor and matches the lower bounds in both regimes n < Θ(K) and n > K of the strongly convex setting, where K denotes the condition number. Besides improving the best known results and matching all the above lower bounds simultaneously, VRADA has more unified and simplified algorithmic implementation and convergence analysis for both the general convex and strongly convex settings. The underlying novel approaches such as the novel initialization strategy in VRADA may be of independent interest. Through experiments on real datasets, we show the good performance of VRADA over existing methods for large-scale machine learning problems. |
Persistent Identifier | http://hdl.handle.net/10722/327773 |
ISSN | 2020 SCImago Journal Rankings: 1.399 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Song, Chaobing | - |
dc.contributor.author | Jinag, Yong | - |
dc.contributor.author | Ma, Yi | - |
dc.date.accessioned | 2023-05-08T02:26:43Z | - |
dc.date.available | 2023-05-08T02:26:43Z | - |
dc.date.issued | 2020 | - |
dc.identifier.citation | Advances in Neural Information Processing Systems, 2020, v. 2020-December | - |
dc.identifier.issn | 1049-5258 | - |
dc.identifier.uri | http://hdl.handle.net/10722/327773 | - |
dc.description.abstract | In this paper, we introduce a simplified and unified method for finite-sum convex optimization, named Variance Reduction via Accelerated Dual Averaging (VRADA). In both general convex and strongly convex settings, VRADA can attain an O(n1)-accurate solution in O(n log log n) number of stochastic gradient evaluations which improves the best known result O(n log n), where n is the number of samples. Meanwhile, VRADA matches the lower bound of the general convex setting up to a log log n factor and matches the lower bounds in both regimes n < Θ(K) and n > K of the strongly convex setting, where K denotes the condition number. Besides improving the best known results and matching all the above lower bounds simultaneously, VRADA has more unified and simplified algorithmic implementation and convergence analysis for both the general convex and strongly convex settings. The underlying novel approaches such as the novel initialization strategy in VRADA may be of independent interest. Through experiments on real datasets, we show the good performance of VRADA over existing methods for large-scale machine learning problems. | - |
dc.language | eng | - |
dc.relation.ispartof | Advances in Neural Information Processing Systems | - |
dc.title | Variance reduction via accelerated dual averaging for finite-sum optimization | - |
dc.type | Conference_Paper | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.scopus | eid_2-s2.0-85108417927 | - |
dc.identifier.volume | 2020-December | - |