Global convergence rate of incremental aggregated gradient methods for nonsmooth problems

Nuri Denizcan Vanli; Mert Gürbüzbalaban; Asuman Ozdaglar

doi:10.1109/CDC.2016.7798265

Conference paper, 2016

Global convergence rate of incremental aggregated gradient methods for nonsmooth problems

Nuri Denizcan Vanli, Mert Gürbüzbalaban, Asuman Ozdaglar

IEEE Conference on Decision and Control (CDC), pp. 173–178, 2016.

Publisher (DOI)

In brief

Extends the linear convergence of incremental aggregated gradient methods to problems with a nonsmooth regularizer, where each step uses one fresh component gradient, stale gradients for the rest, and a proximal step for the regularizer; the rate is explicit in the staleness and the condition number.

Topics

Incremental methods & data ordering

Cite

@inproceedings{vanli2016nonsmooth,
  title   = {Global convergence rate of incremental aggregated gradient methods for nonsmooth problems},
  author  = {Nuri Denizcan Vanli and Mert Gürbüzbalaban and Asuman Ozdaglar},
  year    = {2016},
  booktitle = {IEEE Conference on Decision and Control (CDC)},
  pages   = {173--178},
  doi     = {10.1109/CDC.2016.7798265},
}