English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Understanding the bottlenecks in implementing stochastic gradient descent(SGD)-based distributed support vector machines (SVM) algorithm is important intraining larger data sets. The communication time to do the modelsynchronization across the parallel processes is the main bottleneck thatcauses inefficiency in the training process. The model synchronization isdirectly affected by the mini-batch size of data processed before the globalsynchronization. In producing an efficient distributed model, the communicationtime in training model synchronization has to be as minimum as possible whileretaining a high testing accuracy. The effect from model synchronizationfrequency over the convergence of the algorithm and accuracy of the generatedmodel must be well understood to design an efficient distributed model. In thisresearch, we identify the bottlenecks in model synchronization in parallelstochastic gradient descent (PSGD)-based SVM algorithm with respect to thetraining model synchronization frequency (MSF). Our research shows that byoptimizing the MSF in the data sets that we used, a reduction of 98\% incommunication time can be gained (16x - 24x speed up) with respect tohigh-frequency model synchronization. The training model optimization discussedin this paper guarantees a higher accuracy than the sequential algorithm alongwith faster convergence.

Performance Optimization on Model Synchronization in Parallel Stochastic Gradient Descent Based SVM