Asynchronous Methods in Gradient Descent

Please use this identifier to cite or link to this item: http://hdl.handle.net/10263/7303

Full metadata record

DC Field	Value	Language
dc.contributor.author	Ghosh, Koushik	-
dc.date.accessioned	2022-03-24T05:01:58Z	-
dc.date.available	2022-03-24T05:01:58Z	-
dc.date.issued	2021-07	-
dc.identifier.citation	33p.	en_US
dc.identifier.uri	http://hdl.handle.net/10263/7303	-
dc.description	Dissertation under the supervision of Swagatam Das	en_US
dc.description.abstract	Su, Boyd and Candes ’14 [1] showed that if we make the stepsizes smaller and smaller, Nesterov Accelerated Gradient Descent converges to a 2nd order ODE. On the other hand, arjevani has shown recently some convergence results on delayed vanilla Gradient descent . Our idea is to take a delayed version of Nesterov Accelerated Gradient Descent and derive it’s corresponding ODE and prove convergence for the convex case.	en_US
dc.language.iso	en	en_US
dc.publisher	Indian Statistical Institute, Kolkata.	en_US
dc.relation.ispartofseries	Dissertation;CS-1906	-
dc.subject	Nesterov Accelerated Gradient Descent	en_US
dc.subject	Asynchronous	en_US
dc.subject	Hogwild	en_US
dc.subject	DownPour SGD	en_US
dc.title	Asynchronous Methods in Gradient Descent	en_US
dc.type	Other	en_US
Appears in Collections:	Dissertations - M Tech (CS)

Files in This Item:

File	Description	Size	Format
Koushik Ghosh-cs-19-21.pdf		659.03 kB	Adobe PDF	View/Open