EIFL: Efficiently Protecting Global Model Privacy and Integrity Against an Untrusted Server in Federated Learning

Name: EIFL: Efficiently Protecting Global Model Privacy and Integrity Against an Untrusted Server in Federated Learning
Uploaded: 2026-10-11T15:42:30.354593+00:00
Duration: 400 s
Description: EIFL proposes a novel federated learning method that uses two-stage aggregation and symmetric encryption to protect model output privacy from untrusted servers. It introduces an efficient integrity check using a vector inner product and random vectors, ensuring the final global model is accurate and untampered, while maintaining low communication overhead.

EIFL proposes a novel federated learning method that uses two-stage aggregation and symmetric encryption to protect model output privacy from untrusted servers. It introduces an efficient integrity check using a vector inner product and random vectors, ensuring the final global model is accurate and untampered, while maintaining low communication overhead.

Open the full episode