Secure Embedding Aggregation for Federated Representation Learning

TL;DR

This paper introduces extsc{SecureEmbed}, a privacy-preserving protocol for aggregating local embeddings in federated learning, ensuring privacy against a curious server and colluding clients, enhancing secure collaborative representation learning.

Contribution

The paper proposes extsc{SecureEmbed}, a novel secure aggregation protocol that guarantees privacy for local embeddings in federated learning with collusion resistance.

Findings

Provides privacy guarantees against a curious server and colluding clients.

Leverages all potential aggregation opportunities among clients.

Ensures privacy of local embeddings simultaneously at each client.

Abstract

We consider a federated representation learning framework, where with the assistance of a central server, a group of $N$ distributed clients train collaboratively over their private data, for the representations (or embeddings) of a set of entities (e.g., users in a social network). Under this framework, for the key step of aggregating local embeddings trained privately at the clients, we develop a secure embedding aggregation protocol named \scheme, which leverages all potential aggregation opportunities among all the clients, while providing privacy guarantees for the set of local entities and corresponding embeddings \emph{simultaneously} at each client, against a curious server and up to $T < N/2$ colluding clients.