Friday, 26th July - 11am Joyce Jiyoung Whang : Seminar

Title: Knowledge Graph Representation Learning with Multimodality, Structure, and Theoretical Grounds  

Abstract:

Knowledge graphs represent human knowledge using triplets composed of entities and relations. Knowledge graph representation learning (KGRL) aims to convert the entities and relations into a set of feature vectors while preserving the inherent relationships between them. KGRL is considered one of the most crucial building blocks in many modern AI and machine learning applications. This talk will introduce the latest research exploring KGRL from three perspectives: (1) how can we integrate visual knowledge into the standard textual knowledge graphs, and how can we utilize the visual and textual information in learning representations, (2) how can we fully utilize the structure of a given knowledge graph, and what are the advantages of learning representations using the structure, and (3) how can we show KGRL's generalizability from a theoretical point of view, and how do those theoretical observations guide the design of practical KGRL methods. 

Bio:

Joyce Jiyoung Whang is an associate professor at the School of Computing at KAIST, where she has led the Big Data Intelligence Lab since July 2020. She received her Ph.D. degree in Computer Science from the University of Texas at Austin in December 2015 under the supervision of Professor Inderjit S. Dhillon. She was an assistant professor of Computer Science and Engineering at Sungkyunkwan University (SKKU) from March 2016 to June 2020, leading the Big Data Lab. Her main research interests are graph machine learning, data mining, and big data analytics. In particular, she focuses on developing novel computational algorithms for graph models in various fields; her recent research interests include knowledge graph representation learning and graph neural networks.