THE Industrial Revolution starting from about 1760 and ending at around 1840 has been viewed as the first Industrial Revolution.It features with the replacement of human and animal muscle power with steam and mechanic...THE Industrial Revolution starting from about 1760 and ending at around 1840 has been viewed as the first Industrial Revolution.It features with the replacement of human and animal muscle power with steam and mechanical power.Human income per capita had taken 800 years to double by展开更多
ANFUSI Town in Yichang City,northwest of Zhijiang City,is a national demonstration town of urbanization and rural development in Hubei and site of a large industrial park.The reporter went there to see the work in pro...ANFUSI Town in Yichang City,northwest of Zhijiang City,is a national demonstration town of urbanization and rural development in Hubei and site of a large industrial park.The reporter went there to see the work in progress on building a 4,000-square-meter smart greenhouse with a bucolic landscape and an aquatic product factory equipped to produce surimi.In 2000,there were just a few small local businesses in Anfusi.展开更多
Disk failure prediction methods have been useful in handing a single issue,e.g.,heterogeneous disks,model aging,and minority samples.However,because these issues often exist simultaneously,prediction models that can h...Disk failure prediction methods have been useful in handing a single issue,e.g.,heterogeneous disks,model aging,and minority samples.However,because these issues often exist simultaneously,prediction models that can handle only one will result in prediction bias in reality.Existing disk failure prediction methods simply fuse various models,lacking discussion of training data preparation and learning patterns when facing multiple issues,although the solutions to different issues often conflict with each other.As a result,we first explore the training data preparation for multiple issues via a data partitioning pattern,i.e.,our proposed multi-property data partitioning(MDP).Then,we consider learning with the partitioned data for multiple issues as learning multiple tasks,and introduce the model-agnostic meta-learning(MAML)framework to achieve the learning.Based on these improvements,we propose a novel disk failure prediction model named MDP-MAML.MDP addresses the challenges of uneven partitioning and difficulty in partitioning by time,and MAML addresses the challenge of learning with multiple domains and minor samples for multiple issues.In addition,MDP-MAML can assimilate emerging issues for learning and prediction.On the datasets reported by two real-world data centers,compared to state-of-the-art methods,MDP-MAML can improve the area under the curve(AUC)and false detection rate(FDR)from 0.85 to0.89 and from 0.85 to 0.91,respectively,while reducing false alarm rate(FAR)from 4.88%to 2.85%.展开更多
文摘THE Industrial Revolution starting from about 1760 and ending at around 1840 has been viewed as the first Industrial Revolution.It features with the replacement of human and animal muscle power with steam and mechanical power.Human income per capita had taken 800 years to double by
文摘ANFUSI Town in Yichang City,northwest of Zhijiang City,is a national demonstration town of urbanization and rural development in Hubei and site of a large industrial park.The reporter went there to see the work in progress on building a 4,000-square-meter smart greenhouse with a bucolic landscape and an aquatic product factory equipped to produce surimi.In 2000,there were just a few small local businesses in Anfusi.
基金Project supported by the National Natural Science Foundation of China(No.61902135)the Shandong Provincial Natural Science Foundation,China(No.ZR2019LZH003)。
文摘Disk failure prediction methods have been useful in handing a single issue,e.g.,heterogeneous disks,model aging,and minority samples.However,because these issues often exist simultaneously,prediction models that can handle only one will result in prediction bias in reality.Existing disk failure prediction methods simply fuse various models,lacking discussion of training data preparation and learning patterns when facing multiple issues,although the solutions to different issues often conflict with each other.As a result,we first explore the training data preparation for multiple issues via a data partitioning pattern,i.e.,our proposed multi-property data partitioning(MDP).Then,we consider learning with the partitioned data for multiple issues as learning multiple tasks,and introduce the model-agnostic meta-learning(MAML)framework to achieve the learning.Based on these improvements,we propose a novel disk failure prediction model named MDP-MAML.MDP addresses the challenges of uneven partitioning and difficulty in partitioning by time,and MAML addresses the challenge of learning with multiple domains and minor samples for multiple issues.In addition,MDP-MAML can assimilate emerging issues for learning and prediction.On the datasets reported by two real-world data centers,compared to state-of-the-art methods,MDP-MAML can improve the area under the curve(AUC)and false detection rate(FDR)from 0.85 to0.89 and from 0.85 to 0.91,respectively,while reducing false alarm rate(FAR)from 4.88%to 2.85%.
