A Comparative Study of Federated Learning Aggregation Strategies under Homogeneous and Heterogeneous Data Distributions

TL;DR

This paper compares various federated learning aggregation strategies through extensive experiments on image classification datasets, analyzing their performance, robustness, and efficiency under different data distribution scenarios.

Contribution

It provides a comprehensive experimental analysis of aggregation strategies in federated learning across homogeneous and heterogeneous data distributions, highlighting their trade-offs.

Findings

Aggregation strategies show distinct performance trade-offs depending on data distribution.

Data heterogeneity significantly affects the effectiveness of different aggregation mechanisms.

System efficiency varies with the choice of aggregation strategy and dataset characteristics.

Abstract

Federated Learning has emerged as a transformative paradigm for collaborative machine learning across distributed environments. However, its performance is strongly influenced by the aggregation strategy used to combine local model updates at the server, which directly affects learning performance, robustness, and system behavior. This work presents a comprehensive experimental comparison of widely used federated aggregation strategies under both homogeneous and heterogeneous data distributions. Using benchmark image classification datasets, we analyze how different aggregation mechanisms respond to varying degrees of data heterogeneity, examining their impact on centralized accuracy and loss, and system-level efficiency metrics, including aggregation, training, and communication time. The results demonstrate that aggregation strategies exhibit distinct trade-offs across datasets and data distributions, with their effectiveness varying according to dataset characteristics and operating conditions.