The rapid growth of Internet of things devices and the emergence of rapidly evolving network threats have made traditional security assessment methods inadequate.Federated learning offers a promising solution to exped...The rapid growth of Internet of things devices and the emergence of rapidly evolving network threats have made traditional security assessment methods inadequate.Federated learning offers a promising solution to expedite the training of security assessment models.However,ensuring the trustworthiness and robustness of federated learning under multi-party collaboration scenarios remains a challenge.To address these issues,this study proposes a shard aggregation network structure and a malicious node detection mechanism,along with improvements to the federated learning training process.First,we extract the data features of the participants by using spectral clustering methods combined with a Gaussian kernel function.Then,we introduce a multi-objective decision-making approach that combines data distribution consistency,consensus communication overhead,and consensus result reliability in order to determine the final network sharing scheme.Finally,by integrating the federated learning aggregation process with the malicious node detection mechanism,we improve the traditional decentralized learning process.Our proposed ShardFed algorithm outperforms conventional classification algorithms and state-of-the-art machine learning methods like FedProx and FedCurv in convergence speed,robustness against data interference,and adaptability across multiple scenarios.Experimental results demonstrate that the proposed approach improves model accuracy by up to 2.33%under non-independent and identically distributed data conditions,maintains higher performance with malicious nodes containing poisoned data ratios of 20%–50%,and significantly enhances model resistance to low-quality data.展开更多
The doping effects of transition metals(TMs = Mn, Co, Ni, and Cu) on the superconducting critical parameters are investigated in the films of iron selenide(Li,Fe)OHFe Se. The samples are grown via a matrix-assisted hy...The doping effects of transition metals(TMs = Mn, Co, Ni, and Cu) on the superconducting critical parameters are investigated in the films of iron selenide(Li,Fe)OHFe Se. The samples are grown via a matrix-assisted hydrothermal epitaxy method. Among the TMs, the elements of Mn and Co adjacent to Fe are observed to be incorporated into the crystal lattice more easily. It is suggested that the doped TMs mainly occupy the iron sites of the intercalated(Li,Fe)OH layers rather than those of the superconducting Fe Se layers. We find that the critical current density J_(c) can be enhanced much more strongly by the Mn dopant than the other TMs, while the critical temperature T_(c) is weakly affected by the TM doping.展开更多
The temperature dependences of upper critical field(Hc2) for a series of iron-deficient Fe1-xSe single crystals are obtained from the measurements of in-plane resistivity in magnetic fields up to 9 T and perpendicular...The temperature dependences of upper critical field(Hc2) for a series of iron-deficient Fe1-xSe single crystals are obtained from the measurements of in-plane resistivity in magnetic fields up to 9 T and perpendicular to the ab plane. For the samples with lower superconducting transition temperature Tc(< 7.2 K), the temperature dependence of Hc2 is appropriately described by an effective two-band model. For the samples with higher Tc( 7.2 K), the temperature dependence can also be fitted by a single-band Werthamer–Helfand–Hohenberg formula, besides the two-band model. Such a Tc-dependent change in Hc2(T) behavior is discussed in connection with recent related experimental results, showing an inherent link between the changes of intrinsic superconducting and normal state properties in the Fe Se system.展开更多
基金supported by State Grid Hebei Electric Power Co.,Ltd.Science and Technology Project,Research on Security Protection of Power Services Carried by 4G/5G Networks(Grant No.KJ2024-127).
文摘The rapid growth of Internet of things devices and the emergence of rapidly evolving network threats have made traditional security assessment methods inadequate.Federated learning offers a promising solution to expedite the training of security assessment models.However,ensuring the trustworthiness and robustness of federated learning under multi-party collaboration scenarios remains a challenge.To address these issues,this study proposes a shard aggregation network structure and a malicious node detection mechanism,along with improvements to the federated learning training process.First,we extract the data features of the participants by using spectral clustering methods combined with a Gaussian kernel function.Then,we introduce a multi-objective decision-making approach that combines data distribution consistency,consensus communication overhead,and consensus result reliability in order to determine the final network sharing scheme.Finally,by integrating the federated learning aggregation process with the malicious node detection mechanism,we improve the traditional decentralized learning process.Our proposed ShardFed algorithm outperforms conventional classification algorithms and state-of-the-art machine learning methods like FedProx and FedCurv in convergence speed,robustness against data interference,and adaptability across multiple scenarios.Experimental results demonstrate that the proposed approach improves model accuracy by up to 2.33%under non-independent and identically distributed data conditions,maintains higher performance with malicious nodes containing poisoned data ratios of 20%–50%,and significantly enhances model resistance to low-quality data.
基金Project supported by the National Key Research and Development Program of China(Grant Nos.2017YFA0303003 and 2016YFA0300300)the National Natural Science Foundation of China(Grant Nos.11834016 and 11888101)+1 种基金the Strategic Priority Research Program of Chinese Academy of Sciences(Grant Nos.XDB33010200 and XDB25000000)the Strategic Priority Research Program and Key Research Program of Frontier Sciences of the Chinese Academy of Sciences(Grant Nos.QYZDY-SSW-SLH001 and QYZDY-SSW-SLH008)。
文摘The doping effects of transition metals(TMs = Mn, Co, Ni, and Cu) on the superconducting critical parameters are investigated in the films of iron selenide(Li,Fe)OHFe Se. The samples are grown via a matrix-assisted hydrothermal epitaxy method. Among the TMs, the elements of Mn and Co adjacent to Fe are observed to be incorporated into the crystal lattice more easily. It is suggested that the doped TMs mainly occupy the iron sites of the intercalated(Li,Fe)OH layers rather than those of the superconducting Fe Se layers. We find that the critical current density J_(c) can be enhanced much more strongly by the Mn dopant than the other TMs, while the critical temperature T_(c) is weakly affected by the TM doping.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.11888101 and 11834016)the National Key Research and Development Program of China(Grant Nos.2017YFA0303003 and 2016YFA0300300)the Strategic Priority Research Program and Key Research Program of Frontier Sciences of the Chinese Academy of Sciences(Grant Nos.QYZDY-SSW-SLH001 and XDB25000000)
文摘The temperature dependences of upper critical field(Hc2) for a series of iron-deficient Fe1-xSe single crystals are obtained from the measurements of in-plane resistivity in magnetic fields up to 9 T and perpendicular to the ab plane. For the samples with lower superconducting transition temperature Tc(< 7.2 K), the temperature dependence of Hc2 is appropriately described by an effective two-band model. For the samples with higher Tc( 7.2 K), the temperature dependence can also be fitted by a single-band Werthamer–Helfand–Hohenberg formula, besides the two-band model. Such a Tc-dependent change in Hc2(T) behavior is discussed in connection with recent related experimental results, showing an inherent link between the changes of intrinsic superconducting and normal state properties in the Fe Se system.
