bet365 casino blackjack time：6month2No.（Thursday）Morning10：00-11:30
Tencent conference number：807-949-087
Title:GNNWhen is it valid?
Abstract:Graph Neural Network (GNNs) Learning in graph representationshown inThe powerfulLearning ability, pushedIts practical applications in many fields，includesBiology andHealthMedicaletc.. Therefore, in recent yearsThere are a lot of related onesResearch,bet365 casino blackjack circles have put forward many ideasGNNsModel. However, realityWorldThe picture is very diverse,Existing research stillLack of pairGNNsWhencaiCan be herePractical applicationplaying a role inProfoundUnderstand. This bet365 casino blackjack will include manyGNNsModelRelated to the problem of graph denoising，Provides a new perspective to understandGNNsWhy it works on homogeneous graphs。At the same time，Passed experimentDiscovered,GNNsA lot can be obtained on some commonly used heterogeneous (or non-homogeneous) graphsOKPerformance. further,bet365 casino blackjack willDiscuss homogeneity andNotGNNGood performanceGuaranteenecessary conditions, and willgivenGNNWhen will it be possiblevalidWorkingoneMore general assumptions to support theoretical understanding and empirical observations.
Personal introduction:Tang JiliangThe doctor isAssociate bet365 casino blackjack, Department of Computer Science and Engineering, MSU。bet365 casino blackjack TangWas a scientist at Yahoo Research.himYu2015Received PhD from Arizona State bet365 casino blackjack,2010Nianhe2008respectively。Research directions include graph machine learning、Trustable artificial intelligence and its applications in education and biologyField. Received6numberWell-known in the fieldCareer Award（IBM/SIAM 2022, 2021 IEEE Big Data Security, IEEE ICDM 2021, the Association of Chinese Scholars in Computing 2021, ACM KDD2020, NSF 2019), and8best paper awards (or runner-up), includingWSDM2018andKDD2016。bet365 casino blackjack Tang has participated in organizing many top international eventsConference (such asKDD、SIGIR、WSDMandSDM），Also serves on the editorial board of several top journals(such asTKDDandTKDE). Published in top journals, conferencesMany high-level papers, paperAccumulated by Google ScholarQuote210,000Yutimes,h-indexfor70。himPublished in partnership with Cambridge bet365 casino blackjack PressOne bookAboutGNN's monograph"Graph Deep Learning".
