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

Modern mobile applications are benefiting significantly from the advancementin deep learning, e.g., implementing real-time image recognition andconversational system. Given a trained deep learning model, applicationsusually need to perform a series of matrix operations based on the input data,in order to infer possible output values. Because of computational complexityand size constraints, these trained models are often hosted in the cloud. Toutilize these cloud-based models, mobile apps will have to send input data overthe network. While cloud-based deep learning can provide reasonable responsetime for mobile apps, it restricts the use case scenarios, e.g. mobile appsneed to have network access. With mobile specific deep learning optimizations,it is now possible to employ on-device inference. However, because mobilehardware, such as GPU and memory size, can be very limited when compared to itsdesktop counterpart, it is important to understand the feasibility of this newon-device deep learning inference architecture. In this paper, we empiricallyevaluate the inference performance of three Convolutional Neural Networks(CNNs) using a benchmark Android application we developed. Our measurement andanalysis suggest that on-device inference can cost up to two orders ofmagnitude greater response time and energy when compared to cloud-basedinference, and that loading model and computing probability are two performancebottlenecks for on-device deep inferences.

Cloud-based or On-device: An Empirical Study of Mobile Deep Inference