Threshold-adaptive pruning with multi-key homomorphic encryption for communication-efficient secure federated learning
Jie Guo, Renjing Liu, Jinsheng Xing, Je Sen Teh, Je Sen Teh, Je Sen Teh · 2026
Read Threshold-adaptive pruning with multi-key homomorphic encryption for communication-efficient secure federated learning by Jie Guo, Renjing Liu, Jinsheng Xing, Je Sen Teh, Je Sen Teh, Je Sen Teh (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13183220.txt.
Description
Jie Guo, Renjing Liu, and Jinsheng Xing propose a secure federated learning scheme that cuts communication cost in a 2026 paper. In federated learning, edge devices trade model updates with a server often, which wastes bandwidth. Standard encryption also fails when many nodes collude. Their scheme joins adaptive pruning of model channels with multi key encryption.
A dynamic threshold sets pruning rates from accuracy feedback, balancing compression against quality and cutting bandwidth sharply. A distributed key protocol lets the server merge encrypted updates from many sources without decrypting them, and it resists collusion by all but one node. Tests on two image data sets showed large savings in communication.
Catalog record
| CREATOR | Jie Guo, Renjing Liu, Jinsheng Xing, Je Sen Teh, Je Sen Teh, Je Sen Teh |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 109 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |