Joint Graph Estimation and Signal Restoration for Robust Federated Learning

TL;DR

This paper introduces a robust federated learning aggregation method that learns a graph of client relationships and restores model signals, significantly improving accuracy under noisy communication conditions.

Contribution

It presents a novel joint graph learning and signal restoration approach for robust federated learning, formulated as a difference-of-convex optimization problem.

Findings

Outperforms existing methods by 2-5% in accuracy on MNIST and CIFAR-10.

Effectively handles noisy and biased data distributions.

Demonstrates robustness in communication noise scenarios.

Abstract

We propose a robust aggregation method for model parameters in federated learning (FL) under noisy communications. FL is a distributed machine learning paradigm in which a central server aggregates local model parameters from multiple clients. These parameters are often noisy and/or have missing values during data collection, training, and communication between the clients and server. This may cause a considerable drop in model accuracy. To address this issue, we learn a graph that represents pairwise relationships between model parameters of the clients during aggregation. We realize it with a joint problem of graph learning and signal (i.e., model parameters) restoration. The problem is formulated as a difference-of-convex (DC) optimization, which is efficiently solved via a proximal DC algorithm. Experimental results on MNIST and CIFAR-10 datasets show that the proposed method outperforms existing approaches by up to $2$--$5\%$ in classification accuracy under biased data distributions and noisy conditions.