Traditional federated learning addresses the data security issues arising from the need to centralize client datasets on a central server for model training. However, this approach still poses privacy protection risks. For instance, central servers cannot verify privacy leaks resulting from poisoning attacks by malicious clients. Additionally, adversarial sample attacks can infer specific samples from the original data by testing the local models on client devices. This paper proposes a federated learning privacy protection method combining distillation defense technology with blockchain architecture. The method utilizes distillation defense technology to reduce the sensitivity of client devices participating in federated learning to perturbations and enhance their ability to resist adversarial sample attacks locally. This not only reduces communication overhead and improves learning efficiency but also enhances the model’s generalization ability. Furthermore, the method leverages the “decentralized” nature of blockchain architecture as a trusted record-keeping mechanism to audit information interactions among clients and shared model parameters. This addresses privacy leakage issues resulting from poisoning attacks by some clients during the model construction process. Simulation experiment results demonstrate that the proposed method, compared with traditional federated learning, ensures model convergence, detects malicious clients, and improves the participation level of highly reputable clients. Moreover, by reducing the sensitivity of local clients to perturbations, it enhances their ability to effectively resist adversarial sample attacks.
With the rise of blockchain technology and the Ethereum platform, non-fungible tokens (NFTs) have emerged as a new class of digital assets. The NFT transfer network exhibits core–periphery structures derived from different partitioning methods, leading to local discrepancies and global diversity. We propose a core–periphery structure characterization method based on Bayesian and stochastic block models (SBMs). This method incorporates prior knowledge to improve the fit of core–periphery structures obtained from various partitioning methods. Additionally, we introduce a locally weighted core–periphery structure aggregation (LWCSA) scheme, which determines local aggregation weights using the minimum description length (MDL) principle. This approach results in a more accurate and representative core–periphery structure. The experimental results indicate that core nodes in the NFT transfer network constitute approximately 2.3–5% of all nodes. Compared to baseline methods, our approach improves the normalized mutual information (NMI) index by 6–10%, demonstrating enhanced structural representation. This study provides a theoretical foundation for further analysis of the NFT market.
17 March 2026 in Journal of Network and Systems Management