To study The MALDI-TOF- Mass Spectfometry of silver phenylacetylide in order to investigate The state of aggregation of The functionalized transition metal coordinated compounds including clusters and determine The de...To study The MALDI-TOF- Mass Spectfometry of silver phenylacetylide in order to investigate The state of aggregation of The functionalized transition metal coordinated compounds including clusters and determine The degree of ploymerization of the Silver phenylacetylide. Methods: The silver phenylacetylides were analyzed by The method of MALDI-TOF-MS. Results: The main members of The silver phenylacetylide were tetramer and pentamer The degree of polymerization was found to be up to twelve. Conclusion: The polymerization series of The silver phenylacetylide are according to The formula: Mn= nM+ Ag, (n = 1.2.3... ).展开更多
文摘To study The MALDI-TOF- Mass Spectfometry of silver phenylacetylide in order to investigate The state of aggregation of The functionalized transition metal coordinated compounds including clusters and determine The degree of ploymerization of the Silver phenylacetylide. Methods: The silver phenylacetylides were analyzed by The method of MALDI-TOF-MS. Results: The main members of The silver phenylacetylide were tetramer and pentamer The degree of polymerization was found to be up to twelve. Conclusion: The polymerization series of The silver phenylacetylide are according to The formula: Mn= nM+ Ag, (n = 1.2.3... ).
文摘传统联邦学习训练模型时假定所有参与方可信,但实际场景存在恶意参与方或恶意攻击模型,现有的联邦学习算法面对投毒攻击时,存在模型性能严重下降的问题。针对模型投毒问题,本文提出一种基于联邦平均(federated averaging,Fedavg)与异常检测的联邦检测算法——FedavgCof,该算法考虑到所有参与方之间的差异对比,在中心服务器和本地模型之间添加异常检测层,通过基于聚类的本地异常检测因子(cluster-based local outlier factor,COF)异常检测算法剔除影响模型性能的异常参数,提升模型鲁棒性。实验结果表明,虽然新型投毒方式攻击性更强,但是FedavgCof能够有效防御投毒攻击,降低模型性能损失,提高模型抗投毒攻击能力,相较于Median和模型清洗算法平均提升精度达到10%以上,大幅提升了模型的安全性。