کنفرانس بین المللی محاسبات نرم

صفحه اصلی / ششمین کنفرانس بین المللی محاسبات نرم

Comprehensive Fairness Analysis in Federated Learning with Green Anaconda Optimization: Evaluating Fair Performance in IoT Environments

نویسندگان :

Elahe Eslami¹ Seyed Abolfazl Shahzadeh Fazeli² Jamshid Abouei³ Elham Abbasi⁴

1- Yazd University 2- Yazd University 3- Yazd University 4- Yazd University

کلمات کلیدی :

Federated Learning,،Client Fairness،Green Anaconda Optimization،Metaheuristic Algorithm،Internet of Things،Non-IID Data

چکیده :

Federated Learning (FL) enables the distributed training of AI models on client devices while preserving data privacy. Client fairness remains a critical challenge in FL, especially in heterogeneous Internet of Things (IoT) environments with non-Independent and Identically Distributed (non-IID) data. This paper presents a comprehensive fairness analysis of the Federated Green Anaconda Optimizer (FedGAO) method. Through extensive experiments on CIFAR-10 with non-IID distribution, we evaluate multiple fairness metrics, including Jain Index, Gini Coefficient, accuracy disparity, and statistical parity. Results show FedGAO achieves near-perfect fairness (Jain Index: 1.0000) and reduces accuracy disparity from 10.78% to 1.16% (89% improvement). The method attains 99.57% mean client accuracy, demonstrating uniform performance across heterogeneous clients. Additionally, FedGAO achieves 12.50% higher global accuracy and 80% lower communication costs than FedAvg. These findings establish FedGAO as an efficient approach that excels in accuracy, communication efficiency, and client fairness for Edge-IoT applications.