Complex graph neural networks and cognitive reasoning

  • A+

:金弟(天津大学)
:2023-02-08 15:00
:腾讯会议ID:5799224321(无密码)

报告人:金弟(天津大学)

时  间:202328日下午15:00-16:30

地  点:腾讯会议ID:5799224321(无密码)

内容摘要:

Graph neural network (GNN) has quickly gained the favor of academic and industrial communities since it was proposed, and has become a hot topic in AI research. This report firstly introduces the modeling and solving methods of GNN for complex graph structure. It then introduces the GNN design method that makes GNN smarter, and how to introduce knowledge and reasoning. Finally, the application of GNN in e-commerce search and recommendation is introduced.

人简介:

金弟,天津大学智算学部教授,博士生导师。一直从事图机器学习,特别是网络表示学习、社团发现、图神经网络以及电商搜索推荐方面的研究。近五年一作/通讯发表CCF A类论文30余篇,获CCF A类会议WWW 2021最佳论文奖亚军、国际数据挖掘顶会ICDM 2021最佳学生论文奖亚军、中国社会信息处理大会SMP 2021最佳论文奖、中国计算机教育大会2022最佳论文奖、《自动化学报》年度优秀论文奖,担任中科院一区SCI期刊Information Sciences副主编、SSCI期刊Humanities & Social Sciences Communications副主编,CCF A类会议IJCAI 程序委员会Board MemberIJCAI/AAAI 高级程序委员会成员SPC。主持国家自然基金3项、国家重点研发计划子课题2项。获ACM中国天津新兴奖、中国商业联合会科技进步一等奖。

 

联系人:周达