Cost-TrustFL: Cost-Aware Hierarchical Federated Learning with Lightweight Reputation Evaluation across Multi-Cloud

2025-12-23

AI Summary

Multi-cloud federated learning faces three key challenges: non-IID data distributions, insufficient robustness against poisoning attacks, and high cross-cloud communication costs (e.g., egress fees). To address these, we propose a hierarchical federated learning framework that jointly optimizes model accuracy and economic efficiency. Our method introduces a linear-complexity gradient-approximated Shapley value mechanism for lightweight malicious participant reputation scoring; designs the first cost-aware hierarchical aggregation strategy—prioritizing intra-cloud aggregation to drastically reduce egress traffic; and integrates Byzantine-robust aggregation to enhance security. Evaluated on CIFAR-10 and FEMNIST under 30% malicious clients, our approach achieves 86.7% test accuracy while reducing communication costs by 32%. Moreover, it demonstrates strong robustness to varying degrees of non-IIDness and poisoning attack intensity.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Multiagent Systems: Mechanism Design
• Natural Language Processing: Safety and Robustness

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Federated learning across multi-cloud environments faces critical challenges, including non-IID data distributions, malicious participant detection, and substantial cross-cloud communication costs (egress fees). Existing Byzantine-robust methods focus primarily on model accuracy while overlooking the economic implications of data transfer across cloud providers. This paper presents Cost-TrustFL, a hierarchical federated learning framework that jointly optimizes model performance and communication costs while providing robust defense against poisoning attacks. We propose a gradient-based approximate Shapley value computation method that reduces the complexity from exponential to linear, enabling lightweight reputation evaluation. Our cost-aware aggregation strategy prioritizes intra-cloud communication to minimize expensive cross-cloud data transfers. Experiments on CIFAR-10 and FEMNIST datasets demonstrate that Cost-TrustFL achieves 86.7% accuracy under 30% malicious clients while reducing communication costs by 32% compared to baseline methods. The framework maintains stable performance across varying non-IID degrees and attack intensities, making it practical for real-world multi-cloud deployments.

Problem

• Optimizes model performance and communication costs in multi-cloud federated learning.
• Provides robust defense against poisoning attacks from malicious participants.
• Reduces cross-cloud data transfer expenses through cost-aware aggregation strategies.

Innovation

• Hierarchical federated learning optimizes performance and communication costs
• Gradient-based Shapley value reduces complexity for lightweight reputation evaluation
• Cost-aware aggregation prioritizes intra-cloud to minimize cross-cloud transfers