Show simple item record

dc.contributor.authorGeng, Yanyan
dc.contributor.authorLiang, Ru-Ze
dc.contributor.authorLi, Weizhi
dc.contributor.authorWang, Jingbin
dc.contributor.authorLiang, Gaoyuan
dc.contributor.authorXu, Chenhao
dc.contributor.authorWang, Jing Yan
dc.date.accessioned2021-04-15T11:51:50Z
dc.date.available2021-04-15T11:51:50Z
dc.date.issued2017-01-01
dc.identifier.isbn9782875870391
dc.identifier.urihttp://hdl.handle.net/10754/668801
dc.description.abstractIn the machine learning problems, the performance measure is used to evaluate the machine learning models. Recently, the number positive data points ranked at the top positions (Pos@Top) has been a popular performance measure in the machine learning community. In this paper, we propose to learn a convolutional neural network (CNN) model to maximize the Pos@Top performance measure. The CNN model is used to represent the multi-instance data point, and a classifier function is used to predict the label from the its CNN representation. We propose to minimize the loss function of Pos@Top over a training set to learn the filters of CNN and the classifier parameter. The classifier parameter vector is solved by the Lagrange multiplier method, and the filters are updated by the gradient descent method alternately in an iterative algorithm. Experiments over benchmark data sets show that the proposed method outperforms the state-of-the-art Pos@Top maximization methods.
dc.publisheri6doc.com publication
dc.relation.urlhttps://www.elen.ucl.ac.be/Proceedings/esann/esannpdf/es2017-33.pdf
dc.rightsArchived with thanks to i6doc.com publication
dc.titleLearning convolutional neural network to maximize Pos@Top performance measure
dc.typeConference Paper
dc.contributor.departmentMaterial Science and Engineering Program
dc.contributor.departmentPhysical Science and Engineering (PSE) Division
dc.conference.date2017-04-26 to 2017-04-28
dc.conference.name25th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2017
dc.conference.locationBruges, BEL
dc.eprint.versionPost-print
dc.contributor.institutionProvincial Key Laboratory for Computer Information Processing Technology, Soochow University, Suzhou, 215006, China
dc.contributor.institutionSuning Commerce RandD Center USA Inc., Palo Alto, CA, 94304, United States
dc.contributor.institutionInformation Technology Service Center, Intermediate People's Court of Linyi City, Linyi, China
dc.contributor.institutionJiangsu University of Technology, Jiangsu, 213001, China
dc.contributor.institutionNew York University Abu Dhabi, Abu Dhabi, United Arab Emirates
dc.identifier.pages589-594
kaust.personLiang, Ru-Ze
dc.identifier.eid2-s2.0-85069453744


This item appears in the following Collection(s)

Show simple item record