With the rapid development of the Internet of Things (IoT) and edge computing, the scale and complexity of modern networks have increased significantly, driving the demand for distributed fault diagnosis. Federated learning (FL) effectively addresses the issues of data privacy and dispersion by enabling edge devices to collaboratively train models without sharing raw data. However, existing FL-based fault diagnosis methods still encounter the following challenges. Firstly, static aggregation strategies struggle to balance the contributions of heterogeneous clients dynamically. Secondly, traditional local models are unable to effectively decouple sparse high and low-frequency features in fault signals, thereby limiting the accuracy of fault identification. Finally, the resource constraints of edge devices restrict the deployment of complex diagnostic models. To address these challenges, we propose a federated learning hybrid dynamic weight adjustment method based on delay and model quality, introducing the concept of “accelerated depreciation” in accounting and taxation and the concept of “asset allocation” in economics to improve the communication efficiency of fault diagnosis and reduce the impact of delay differences due to device heterogeneity on the effect of fault diagnosis. Additionally, we propose an adaptive sparse low high frequencies Transformer, introducing a lightweight attention mechanism and an adaptive feature extraction layer, which significantly reduces the computational overhead while maintaining high diagnostic accuracy. The experimental results show that, compared with the most competitive baseline, our method improves the fault diagnosis accuracy by 1% on the Case Western Reserve University Bearing dataset (CWRU), 0.78% on the Xi’an Jiaotong University Gearbox dataset (XJTU), and 0.44% on the Micro service Edge Computing dataset (MICRO).
Jingting Mei, Yang Yang, Celimuge Wu et al.· IEEE Transactions on Cogniti...· 0 citations
In agentic AI-enabled edge computing, decentralized federated learning (DFL) leverages peer-to-peer model aggregation to improve the performance of on-device large language models (LLMs) without introducing a single point of failure, thereby enhancing local agents’ capabilities for decision-making. To protect the right to be forgotten for each agent, as required by data regulations such as the General Data Protection Regulation (GDPR), federated unlearning aims to remove the influence of a target agent’s data from the trained LLM while preserving model utility. However, existing federated unlearning methods predominantly assume centralized architectures and face two critical challenges when extended to decentralized federated learning systems: 1) requiring all remaining agents to participate in the unlearning process incurs prohibitive overhead; and 2) updating all Low-Rank Adaptation (LoRA) modules indiscriminately leads to excessive resource consumption while potentially degrading model performance. To tackle these challenges, we propose a dual-level selective unlearning framework (DSU) for decentralized federated learning. At the agent level, DSU selects retained agents by matching historical LoRA update sketches, label sketches, and exposure to the withdrawn agent. Meanwhile, we derive a sensitivity score from the DFL training trajectory that measures each LoRA module’s accumulated influence from the target agent via the energy of its effective weight changes, and then update only the most sensitive modules while freezing the rest. Theoretical analysis relates selective unlearning to full-participation unlearning, and experiments show that DSU reduces the audited target-agent influence while preserving retained utility and reducing both participating agents and updated LoRA modules in the evaluated settings. The public reproducibility package is available at https://github.com/DGL-codes/LLM-DFL
Zhiqiang Xie, Yijing Lin, Zhipeng Gao et al.· IEEE Transactions on Cogniti...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
