Reporter: Yseabet app download Nan UAEMBZUAIPostdocseabet app downloadral fellow
Reportseabet app downloadg location: Computer Buildseabet app downloadg of the school headquarters313Lectuseabet app download Hall
seabet app download time:2023Year11month15Sunday (Wednesday) mornseabet app downloadg10:00
seabet app download title: "Graph Domain Transfer Learning"
Personal seabet app downloadtroduction: Yseabet app download Nan,MBZUAIPostdoc。Studying at National University of Defense Technology for bachelor’s degree, master’s degree and Ph.D.，In machine learning、Social Multimedia、Has deep professional knowledge and years of research experience in areas such as seabet app download neural networks，Achieved research results recognized at home and abroad in related fields。In the past five years，Participate in publishing high-quality papers10More articles, seabet app downloadcludseabet app downloadgCCF AFirst author of journals and confeseabet app downloadnces/Correspondseabet app downloadg author’s paper7Article. seabet app downloadrveAAAI、NeurIPS、ICML、ACMMMand other seabet app downloadp international artificial intelligence conference reviewers, and served asTCSVT、seabet app downloadMM、TCSSReviewer for many seabet app downloadternational journals seabet app download the fields of artificial seabet app downloadtelligence and data mseabet app downloadseabet app downloadg。
Abstract: Although seabet app download neural network (GNN) has achieved impressive achievements in seabet app download classification tasks，But they often require a large number of task-specific tags，And obtaining these tags can be very expensive。A reliable solution is to explore additional labeled graphs to enhance unsupervised learning of the target domain。However，Due to insufficient exploration of seabet app download topology and significant differences between domains，How toGNNIntegration with domain migration is still pending。Therefore，The seabet app download introduces related methods of graph domain transfer learning，These methods conduct in-depth exploration on how to achieve graph domain alignment，Effectively solves the problems related to graph domain transfer learning。
