A Secure and Efficient Federated Learning Framework for NLP
This addresses the need for secure and efficient federated learning in NLP, offering improvements over existing methods but is incremental in nature.
The paper tackled the problem of designing secure and efficient federated learning frameworks by proposing SEFL, which eliminates trusted entities, achieves comparable or better model accuracy, and is resilient to client dropouts, with a pruning technique improving runtime performance up to 13.7x.
In this work, we consider the problem of designing secure and efficient federated learning (FL) frameworks. Existing solutions either involve a trusted aggregator or require heavyweight cryptographic primitives, which degrades performance significantly. Moreover, many existing secure FL designs work only under the restrictive assumption that none of the clients can be dropped out from the training protocol. To tackle these problems, we propose SEFL, a secure and efficient FL framework that (1) eliminates the need for the trusted entities; (2) achieves similar and even better model accuracy compared with existing FL designs; (3) is resilient to client dropouts. Through extensive experimental studies on natural language processing (NLP) tasks, we demonstrate that the SEFL achieves comparable accuracy compared to existing FL solutions, and the proposed pruning technique can improve runtime performance up to 13.7x.