Private Collaborative Edge Inference via Over-the-Air Computation

We consider collaborative inference at the wireless edge, where each client’s model is trained independently on its local dataset. Clients are queried in parallel to make an accurate decision collaboratively. In addition to maximizing the inference accuracy, we also want to ensure the pri...

Full description

Saved in:
Bibliographic Details
Main Authors: Selim F. Yilmaz, Burak Hasircioglu, Li Qiao, Deniz Gunduz
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Transactions on Machine Learning in Communications and Networking
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10829586/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849723147621761024
author Selim F. Yilmaz
Burak Hasircioglu
Li Qiao
Deniz Gunduz
author_facet Selim F. Yilmaz
Burak Hasircioglu
Li Qiao
Deniz Gunduz
author_sort Selim F. Yilmaz
collection DOAJ
description We consider collaborative inference at the wireless edge, where each client’s model is trained independently on its local dataset. Clients are queried in parallel to make an accurate decision collaboratively. In addition to maximizing the inference accuracy, we also want to ensure the privacy of local models. To this end, we leverage the superposition property of the multiple access channel to implement bandwidth-efficient multi-user inference methods. We propose different methods for ensemble and multi-view classification that exploit over-the-air computation (OAC). We show that these schemes perform better than their orthogonal counterparts with statistically significant differences while using fewer resources and providing privacy guarantees. We also provide experimental results verifying the benefits of the proposed OAC approach to multi-user inference, and perform an ablation study to demonstrate the effectiveness of our design choices. We share the source code of the framework publicly on Github to facilitate further research and reproducibility.
format Article
id doaj-art-9f5bdf818ef04549afbf7ad14814dc26
institution DOAJ
issn 2831-316X
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Transactions on Machine Learning in Communications and Networking
spelling doaj-art-9f5bdf818ef04549afbf7ad14814dc262025-08-20T03:11:06ZengIEEEIEEE Transactions on Machine Learning in Communications and Networking2831-316X2025-01-01321523110.1109/TMLCN.2025.352655110829586Private Collaborative Edge Inference via Over-the-Air ComputationSelim F. Yilmaz0https://orcid.org/0000-0002-0486-7731Burak Hasircioglu1https://orcid.org/0000-0001-5005-2894Li Qiao2https://orcid.org/0000-0003-0586-189XDeniz Gunduz3https://orcid.org/0000-0002-7725-395XDepartment of Electrical and Electronic Engineering, Imperial College London, London, U.K.The Alan Turing Institute, London, U.K.School of Information and Electronics, Beijing Institute of Technology, Beijing, ChinaDepartment of Electrical and Electronic Engineering, Imperial College London, London, U.K.We consider collaborative inference at the wireless edge, where each client’s model is trained independently on its local dataset. Clients are queried in parallel to make an accurate decision collaboratively. In addition to maximizing the inference accuracy, we also want to ensure the privacy of local models. To this end, we leverage the superposition property of the multiple access channel to implement bandwidth-efficient multi-user inference methods. We propose different methods for ensemble and multi-view classification that exploit over-the-air computation (OAC). We show that these schemes perform better than their orthogonal counterparts with statistically significant differences while using fewer resources and providing privacy guarantees. We also provide experimental results verifying the benefits of the proposed OAC approach to multi-user inference, and perform an ablation study to demonstrate the effectiveness of our design choices. We share the source code of the framework publicly on Github to facilitate further research and reproducibility.https://ieeexplore.ieee.org/document/10829586/Edge inferencecollaborative inferencedistributed inferenceensemblemulti-viewover-the-air computation (OAC)
spellingShingle Selim F. Yilmaz
Burak Hasircioglu
Li Qiao
Deniz Gunduz
Private Collaborative Edge Inference via Over-the-Air Computation
IEEE Transactions on Machine Learning in Communications and Networking
Edge inference
collaborative inference
distributed inference
ensemble
multi-view
over-the-air computation (OAC)
title Private Collaborative Edge Inference via Over-the-Air Computation
title_full Private Collaborative Edge Inference via Over-the-Air Computation
title_fullStr Private Collaborative Edge Inference via Over-the-Air Computation
title_full_unstemmed Private Collaborative Edge Inference via Over-the-Air Computation
title_short Private Collaborative Edge Inference via Over-the-Air Computation
title_sort private collaborative edge inference via over the air computation
topic Edge inference
collaborative inference
distributed inference
ensemble
multi-view
over-the-air computation (OAC)
url https://ieeexplore.ieee.org/document/10829586/
work_keys_str_mv AT selimfyilmaz privatecollaborativeedgeinferenceviaovertheaircomputation
AT burakhasircioglu privatecollaborativeedgeinferenceviaovertheaircomputation
AT liqiao privatecollaborativeedgeinferenceviaovertheaircomputation
AT denizgunduz privatecollaborativeedgeinferenceviaovertheaircomputation