Mini-Batch Primal and Dual Methods for SVMs


We address the issue of using mini-batches in stochastic optimization of SVMs. We show that the same quantity, the spectral norm of the data, controls the parallelization speedup obtained for both primal stochastic subgradi-ent descent (SGD) and stochastic dual coordinate ascent (SCDA) methods and use it to derive novel variants of mini-batched SDCA. Our guarantees for both methods are expressed in terms of the original nonsmooth primal problem based on the hinge-loss.

Extracted Key Phrases

3 Figures and Tables

Citations per Year

88 Citations

Semantic Scholar estimates that this publication has received between 59 and 137 citations based on the available data.

See our FAQ for additional information.