Input Selection for Bandwidth-Limited Neural Network Inference

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

Dokumenter

  • Fulltext

    Accepteret manuskript, 2,74 MB, PDF-dokument

Data are often accommodated on centralized storage servers. This is the case, for instance, in remote sensing and astronomy, where projects produce several petabytes of data every year. While machine learning models are often trained on relatively small subsets of the data, the inference phase typically requires transferring significant amounts of data between the servers and the clients. In many cases, the bandwidth available per user is limited, which then renders the data transfer to be one of the major bottlenecks. In this work, we propose a framework that automatically selects the relevant parts of the input data for a given neural network. The model as well as the associated selection masks are trained simultaneously such that a good model performance is achieved while only a minimal amount of data is selected. During the inference phase, only those parts of the data have to be transferred between the server and the client. We propose both instance-independent and instance-dependent selection masks. The former ones are the same for all instances to be transferred, whereas the latter ones allow for variable transfer sizes per instance. Our experiments show that it is often possible to significantly reduce the amount of data needed to be transferred without affecting the model quality much.

OriginalsprogEngelsk
TitelProceedings of the 2022 SIAM International Conference on Data Mining, SDM 2022
Antal sider9
ForlagSIAM
Publikationsdato2022
Sider280-288
ISBN (Elektronisk)9781611977172
DOI
StatusUdgivet - 2022
Begivenhed2022 SIAM International Conference on Data Mining, SDM 2022 - Virtual, Online
Varighed: 28 apr. 202230 apr. 2022

Konference

Konference2022 SIAM International Conference on Data Mining, SDM 2022
ByVirtual, Online
Periode28/04/202230/04/2022

Bibliografisk note

Funding Information:
Changes in Big Satellite Data via Massively-Parallel Artificial Intelligence (9131-00110B). We also acknowledge support by the Villum Foundation through the project Deep Learning and Remote Sensing for Unlocking Global Ecosystem Resource Dynamics (DeReEco).

Funding Information:
We acknowledge support from the Independent Research Fund Denmark through the grant Monitoring

Publisher Copyright:
Copyright © 2022 by SIAM.

ID: 314303760