DPFNAS: Differential Privacy-Enhanced Federated Neural Architecture Search for 6G Edge Intelligence

2025-09-26

AI Summary

To address two critical challenges in federated learning (FL) for 6G edge intelligence—data reconstruction risks arising from parameter sharing and poor global model generalization to non-IID local data—this paper proposes a novel FL framework integrating personalized differential privacy (PDP) and adaptive neural architecture search (NAS). First, it pioneers the application of PDP at the representation level, replacing raw parameter exchange with sample-level feature sharing to fundamentally mitigate reconstruction attacks. Second, it introduces a privacy-aware federated NAS mechanism that discovers lightweight, client-specific architectures under rigorous differential privacy constraints. Theoretical analysis establishes convergence guarantees. Experiments on CIFAR-10/100 demonstrate that the method achieves a 6.82% accuracy improvement over PerFedRLNAS, reduces model size to 1/10, and cuts communication overhead to 1/20.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Learning to Search
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Security and privacy of machine learning and AI applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The Sixth-Generation (6G) network envisions pervasive artificial intelligence (AI) as a core goal, enabled by edge intelligence through on-device data utilization. To realize this vision, federated learning (FL) has emerged as a key paradigm for collaborative training across edge devices. However, the sensitivity and heterogeneity of edge data pose key challenges to FL: parameter sharing risks data reconstruction, and a unified global model struggles to adapt to diverse local distributions. In this paper, we propose a novel federated learning framework that integrates personalized differential privacy (DP) and adaptive model design. To protect training data, we leverage sample-level representations for knowledge sharing and apply a personalized DP strategy to resist reconstruction attacks. To ensure distribution-aware adaptation under privacy constraints, we develop a privacy-aware neural architecture search (NAS) algorithm that generates locally customized architectures and hyperparameters. To the best of our knowledge, this is the first personalized DP solution tailored for representation-based FL with theoretical convergence guarantees. Our scheme achieves strong privacy guarantees for training data while significantly outperforming state-of-the-art methods in model performance. Experiments on benchmark datasets such as CIFAR-10 and CIFAR-100 demonstrate that our scheme improves accuracy by 6.82% over the federated NAS method PerFedRLNAS, while reducing model size to 1/10 and communication cost to 1/20.

Problem

• Protecting sensitive edge data from reconstruction attacks
• Adapting models to diverse local data distributions
• Reducing communication costs while maintaining model performance

Innovation

• Personalized differential privacy protects training data
• Privacy-aware NAS generates customized local architectures
• Sample-level representations enable secure knowledge sharing