Towards Robust Federated Image Classification: An Empirical Study of Weight Selection Strategies in Manufacturing

TL;DR

This paper empirically compares weight selection strategies in federated learning for manufacturing image classification, focusing on model robustness and convergence with limited clients and multiple neural architectures.

Contribution

It provides a comparative analysis of FEWS and OEWS strategies in federated learning within manufacturing, highlighting their impact on model performance and robustness.

Findings

OEWS generally outperforms FEWS in model accuracy

Optimal epoch selection improves convergence speed

Results vary across neural network architectures

Abstract

In the realm of Federated Learning (FL), particularly within the manufacturing sector, the strategy for selecting client weights for server aggregation is pivotal for model performance. This study investigates the comparative effectiveness of two weight selection strategies: Final Epoch Weight Selection (FEWS) and Optimal Epoch Weight Selection (OEWS). Designed for manufacturing contexts where collaboration typically involves a limited number of partners (two to four clients), our research focuses on federated image classification tasks. We employ various neural network architectures, including EfficientNet, ResNet, and VGG, to assess the impact of these weight selection strategies on model convergence and robustness. Our research aims to determine whether FEWS or OEWS enhances the global FL model's performance across communication rounds (CRs). Through empirical analysis and rigorous experimentation, we seek to provide valuable insights for optimizing FL implementations in manufacturing, ensuring that collaborative efforts yield the most effective and reliable models with a limited number of participating clients. The findings from this study are expected to refine FL practices significantly in manufacturing, thereby enhancing the efficiency and performance of collaborative machine learning endeavors in this vital sector.