随着联邦学习的不断兴起,梯度提升决策树(GBDT)作为一种传统的机器学习方法,逐渐应用于联邦学习中以达到理想的分类效果。针对现有GBDT的横向联邦学习模型,存在精度受非独立同分布数据的影响较大、信息泄露和通信成本高等问题,提出了一...随着联邦学习的不断兴起,梯度提升决策树(GBDT)作为一种传统的机器学习方法,逐渐应用于联邦学习中以达到理想的分类效果。针对现有GBDT的横向联邦学习模型,存在精度受非独立同分布数据的影响较大、信息泄露和通信成本高等问题,提出了一种面向非独立同分布数据的联邦梯度提升决策树(federated GBDT for non-IID dataset,nFL-GBDT)。首先,采用局部敏感哈希(LSH)来计算各个参与方之间的相似样本,通过加权梯度来构建第一棵树。其次,由可靠第三方计算只需要一轮通信的全局叶权重来更新树模型。最后,实验分析表明了该算法能够实现对原始数据的隐私保护,并且通信成本低于simFL和FederBoost。同时,实验按照不平衡比率来划分三组公共的数据集,结果表明该算法与Individual、TFL及F-GBDT-G相比,准确率分别提升了3.53%、5.46%和4.43%。展开更多
Power systems transport an increasing amount of electricity,and in the future,involve more distributed renewables and dynamic interactions of the equipment.The system response to disturbances must be secure and predic...Power systems transport an increasing amount of electricity,and in the future,involve more distributed renewables and dynamic interactions of the equipment.The system response to disturbances must be secure and predictable to avoid power blackouts.The system response can be simulated in the time domain.However,this dynamic security assessment(DSA)is not computationally tractable in real-time.Particularly promising is to train decision trees(DTs)from machine learning as interpretable classifiers to predict whether the systemwide responses to disturbances are secure.In most research,selecting the best DT model focuses on predictive accuracy.However,it is insufficient to focus solely on predictive accuracy.Missed alarms and false alarms have drastically different costs,and as security assessment is a critical task,interpretability is crucial for operators.In this work,the multiple objectives of interpretability,varying costs,and accuracies are considered for DT model selection.We propose a rigorous workflow to select the best classifier.In addition,we present two graphical approaches for visual inspection to illustrate the selection sensitivity to probability and impacts of disturbances.We propose cost curves to inspect selection combining all three objectives for the first time.Case studies on the IEEE 68 bus system and the French system show that the proposed approach allows for better DT-selections,with an 80%increase in interpretability,5%reduction in expected operating cost,while making almost zero accuracy compromises.The proposed approach scales well with larger systems and can be used for models beyond DTs.Hence,this work provides insights into criteria for model selection in a promising application for methods from artificial intelligence(AI).展开更多
文摘随着联邦学习的不断兴起,梯度提升决策树(GBDT)作为一种传统的机器学习方法,逐渐应用于联邦学习中以达到理想的分类效果。针对现有GBDT的横向联邦学习模型,存在精度受非独立同分布数据的影响较大、信息泄露和通信成本高等问题,提出了一种面向非独立同分布数据的联邦梯度提升决策树(federated GBDT for non-IID dataset,nFL-GBDT)。首先,采用局部敏感哈希(LSH)来计算各个参与方之间的相似样本,通过加权梯度来构建第一棵树。其次,由可靠第三方计算只需要一轮通信的全局叶权重来更新树模型。最后,实验分析表明了该算法能够实现对原始数据的隐私保护,并且通信成本低于simFL和FederBoost。同时,实验按照不平衡比率来划分三组公共的数据集,结果表明该算法与Individual、TFL及F-GBDT-G相比,准确率分别提升了3.53%、5.46%和4.43%。
基金The authors were supported by a scholarship funded by the Nige-rian National Petroleum Corporation,NNPC,the TU Delft AI Labs Programme,NL,and the research project IDLES,UK(EP/R045518/1).
文摘Power systems transport an increasing amount of electricity,and in the future,involve more distributed renewables and dynamic interactions of the equipment.The system response to disturbances must be secure and predictable to avoid power blackouts.The system response can be simulated in the time domain.However,this dynamic security assessment(DSA)is not computationally tractable in real-time.Particularly promising is to train decision trees(DTs)from machine learning as interpretable classifiers to predict whether the systemwide responses to disturbances are secure.In most research,selecting the best DT model focuses on predictive accuracy.However,it is insufficient to focus solely on predictive accuracy.Missed alarms and false alarms have drastically different costs,and as security assessment is a critical task,interpretability is crucial for operators.In this work,the multiple objectives of interpretability,varying costs,and accuracies are considered for DT model selection.We propose a rigorous workflow to select the best classifier.In addition,we present two graphical approaches for visual inspection to illustrate the selection sensitivity to probability and impacts of disturbances.We propose cost curves to inspect selection combining all three objectives for the first time.Case studies on the IEEE 68 bus system and the French system show that the proposed approach allows for better DT-selections,with an 80%increase in interpretability,5%reduction in expected operating cost,while making almost zero accuracy compromises.The proposed approach scales well with larger systems and can be used for models beyond DTs.Hence,this work provides insights into criteria for model selection in a promising application for methods from artificial intelligence(AI).
基金the Australian large ARC grants No.DP0449535DP0559536 and DP0667060(澳大利亚ARC项目)+4 种基金the Key Project of National Natural Science Foundation of China under Grant No.60496327(国家自然科学基金重大项目)the National Natural Science Foundation of China under Grant No.60463003(国家自然科学基金)the Overseas-Returning High-level Talent Research Pro-gram of Human-Resource Ministry of China(人事部(2004-2005年度首批)海外高层次留学人才回国资助项目)the Innovation Project of Guangxi Graduate Education No.2006106020812M35(广西研究生教育创新计划资助项目).
