SCALE: Sensitivity-Aware Federated Unlearning with Information Freshness Optimization for Mobile Edge Computing

2026-05-21

AI Summary

Existing federated unlearning methods in mobile edge computing suffer from insufficient data removal accuracy due to their neglect of information timeliness and coarse-grained unlearning strategies. To address this limitation, this work proposes SCALE, a novel framework that integrates historical contribution assessment with an information freshness-aware mechanism to enable adaptive sparsification at both layer and weight-subgroup granularities, achieving fine-grained and high-precision federated unlearning. SCALE leverages layer-wise sensitivity analysis to guide sparsification policies, effectively balancing model convergence and training efficiency. Experimental results demonstrate that SCALE significantly outperforms existing approaches, enhancing unlearning accuracy while ensuring timely erasure of deleted data.

Abstract

Federated Unlearning (FU) is emerging as a powerful tool that enables the selective removal of client data to effectively address data contamination and meet strict privacy regulations in mobile edge computing (MEC) systems. Although FU has recently drawn attention in the AI community, existing approaches suffer from low unlearning precision and lack temporal information reflection, which results in suboptimal forgetting performance. To address these issues, we propose SCALE, a dual-level unlearning framework combining historical contribution analysis with information freshness-aware adaptive sparsification. Our framework first employs a historical contribution-based layer sensitivity analysis to identify layers most influenced by target clients, then performs fine-grained unlearning through adaptive sparsification at the weight sub-group level to balance information freshness with forgetting effectiveness. Through theoretical analysis, the proposed framework demonstrates the convergence properties and acceleration advantages. Our experiments and testbed results demonstrate superior unlearning effectiveness compared to state-of-the-art baselines, with significantly improved forgetting performance.

Problem

• Federated Unlearning
• Information Freshness
• Mobile Edge Computing
• Unlearning Precision
• Data Contamination

Innovation

• Federated Unlearning
• Sensitivity Analysis
• Information Freshness
• Adaptive Sparsification
• Mobile Edge Computing