DISTRIBUTION OF THE NEURAL NETWORK BETWEEN MOBILE DEVICE AND CLOUD INFRASTRUCTURE SERVICES

Abstract

Neural networks become the only way to solve problems in some areas. Such tasks as recognition of images, sounds, classification require serious processor power and memory for training and functioning of the network. Modern mobile devices have quite good characteristics for primary layers of deep neural networks, but there are not enough resources for whole network. Since neural networks for mobile devices are trained separately on external resources, a method of distributed work of a neural network with vertical distribution over sets of layers with synchronization of training data was developed. The model is divided after saving its state, all layers on the mobile device are converted to the format for the mobile framework and synchronized with the device after training on a distributed platform. Variables and coefficients are formed separately, which allows to significantly reduce the size of the neural network data file uploaded to the device. An algorithm for automatic selection of a neural network separation point was proposed. It based on the data amount transferred between the layers and the load on the mobile device resources. The approach allows to use full-size deep neural networks with a mobile device. Performance experiment showed possibility of obtains an acceptable response even with an unstable communication channel without overloading communication channels and device resources.

Authors and Affiliations

Yuri Ushakov, Petr Polezhaev Petr, Aleksandr Shukhman, Margarita Ushakova

Keywords

Related Articles

RESOURCE INTENSITY CALCULATION For TEXTS LANGUAGE IDENTIFICATION PROGRAMS

The article deals with the resource intensity calculation of texts language identification programs depending on their identification methods. The most commonly used methods are described, together with an indication of...

TRANSPORT AND ECONOMIC BALANCE AND ITS ROLE IN THE COORDINATION OF TRANSPORT PLANNING DURING THE DIGITALIZATION ERA

Econometrics tools in framework of Transport and Economic Balance (TEB) for transport planning and forecasting are aimed at improving transport links and transport connectivity via accounting the ratio between the volume...

SIGN LANGUAGE NOTATION SYSTEMS AND THEIR COMPARATIVE ANALYSIS

In this paper we describe gestures and sign language as means of non-verbal communication of people with hearing loss. We give a brief overview of existing sign language notations on currently, designed to depict gesture...

JINR CLOUD SERVICE FOR SCIENTIFIC AND ENGINEERING COMPUTATIONS

Pretty often small research scientific groups do not have access to powerful enough computational resources required for their research work to be productive. Global computational infrastructures used by large scientific...

SOLUTION OF THE PROBLEM OF HIGH-PRECISION POSITIONING OF AUTOMOBILE TRANSPORT ON THE BASIS OF THE USE OF ELECTRONIC MAPS

The solution of the problem of positioning of moving objects is now increasingly carried out using electronic maps, allowing approximating with high accuracy the trajectory of the object by a set of orthodromic trajector...

Download PDF file
  • EP ID EP523569
  • DOI 10.25559/SITITO.14.201804.903-910
  • Views 88
  • Downloads 0

How To Cite

Yuri Ushakov, Petr Polezhaev Petr, Aleksandr Shukhman, Margarita Ushakova (2018). DISTRIBUTION OF THE NEURAL NETWORK BETWEEN MOBILE DEVICE AND CLOUD INFRASTRUCTURE SERVICES. Современные информационные технологии и ИТ-образование, 14(4), 903-910. https://europub.co.uk/articles/-A-523569