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

This paper investigates task-oriented communication for multi-devicecooperative edge inference, where a group of distributed low-end edge devicestransmit the extracted features of local samples to a powerful edge server forinference. While cooperative edge inference can overcome the limited sensingcapability of a single device, it substantially increases the communicationoverhead and may incur excessive latency. To enable low-latency cooperativeinference, we propose a learning-based communication scheme that optimizeslocal feature extraction and distributed feature encoding in a task-orientedmanner, i.e., to remove data redundancy and transmit information that isessential for the downstream inference task rather than reconstructing the datasamples at the edge server. Specifically, we leverage an information bottleneck(IB) principle to extract the task-relevant feature at each edge device andadopt a distributed information bottleneck (DIB) framework to formalize asingle-letter characterization of the optimal rate-relevance tradeoff fordistributed feature encoding. To admit flexible control of the communicationoverhead, we extend the DIB framework to a distributed deterministicinformation bottleneck (DDIB) objective that explicitly incorporates therepresentational costs of the encoded features. As the IB-based objectives arecomputationally prohibitive for high-dimensional data, we adopt variationalapproximations to make the optimization problems tractable. To compensate thepotential performance loss due to the variational approximations, we alsodevelop a selective retransmission (SR) mechanism to identify the redundancy inthe encoded features of multiple edge devices to attain additionalcommunication overhead reduction. Extensive experiments evidence that theproposed task-oriented communication scheme achieves a better rate-relevancetradeoff than baseline methods.

Task-Oriented Communication for Multi-Device Cooperative Edge Inference