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/

Zendy Plus

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

Zendy Tools

Zendy Open

Consider a device that is connected to an edge processor via a communicationchannel. The device holds local data that is to be offloaded to the edgeprocessor so as to train a machine learning model, e.g., for regression orclassification. Transmission of the data to the learning processor, as well astraining based on Stochastic Gradient Descent (SGD), must be both completedwithin a time limit. Assuming that communication and computation can bepipelined, this letter investigates the optimal choice for the packet payloadsize, given the overhead of each data packet transmission and the ratio betweenthe computation and the communication rates. This amounts to a tradeoff betweenbias and variance, since communicating the entire data set first reduces thebias of the training process but it may not leave sufficient time for learning.Analytical bounds on the expected optimality gap are derived so as to enable aneffective optimization, which is validated in numerical results.

Optimizing Pipelined Computation and Communication for Latency-Constrained Edge Learning