Collaborative Edge Inference via Semantic Grouping under Wireless Channel Constraints
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912623609511936 |
|---|---|
| author | Mota, Mateus P. Merluzzi, Mattia Strinati, Emilio Calvanese |
| author_facet | Mota, Mateus P. Merluzzi, Mattia Strinati, Emilio Calvanese |
| contents | In this paper, we study the framework of collaborative inference, or edge ensembles. This framework enables multiple edge devices to improve classification accuracy by exchanging intermediate features rather than raw observations. However, efficient communication strategies are essential to balance accuracy and bandwidth limitations. Building upon a key-query mechanism for selective information exchange, this work extends collaborative inference by studying the impact of channel noise in feature communication, the choice of intermediate collaboration points, and the communication-accuracy trade-off across tasks. By analyzing how different collaboration points affect performance and exploring communication pruning, we show that it is possible to optimize accuracy while minimizing resource usage. We show that the intermediate collaboration approach is robust to channel errors and that the query transmission needs a higher degree of reliability than the data transmission itself. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_02222 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Collaborative Edge Inference via Semantic Grouping under Wireless Channel Constraints Mota, Mateus P. Merluzzi, Mattia Strinati, Emilio Calvanese Information Theory Signal Processing In this paper, we study the framework of collaborative inference, or edge ensembles. This framework enables multiple edge devices to improve classification accuracy by exchanging intermediate features rather than raw observations. However, efficient communication strategies are essential to balance accuracy and bandwidth limitations. Building upon a key-query mechanism for selective information exchange, this work extends collaborative inference by studying the impact of channel noise in feature communication, the choice of intermediate collaboration points, and the communication-accuracy trade-off across tasks. By analyzing how different collaboration points affect performance and exploring communication pruning, we show that it is possible to optimize accuracy while minimizing resource usage. We show that the intermediate collaboration approach is robust to channel errors and that the query transmission needs a higher degree of reliability than the data transmission itself. |
| title | Collaborative Edge Inference via Semantic Grouping under Wireless Channel Constraints |
| topic | Information Theory Signal Processing |
| url | https://arxiv.org/abs/2510.02222 |