Talks and presentations
See a map of all the places I've given a talk!
December 15, 2019
Guest Lecture, Introduction to Artificial Intelligence, School of Software, Nankai University, Tianjin, China
In the last decade, deep learning has been a ‘crown jewel’ in artificial intelligence and machine learning. However, utilizing deep learning methods for analyzing the ubiquitous graph data is a non-trivial problem, which attracted considerable research attention in the past few years. In this guest lecture, I will present the categorization of graph-based deep learning methods and review these methods following their history of developments and how these methods solve challenges of graphs. Finally, I will discuss potential future directions. More details can be found in our survey paper: https://arxiv.org/abs/1812.04202.
March 15, 2019
Talk, S-Lab, School of Computer Science and Engineering, Beihang University, Beijing, China
In the last decade, deep learning has been a ‘crown jewel’ in artificial intelligence and machine learning. However, utilizing deep learning methods for analyzing the ubiquitous graph data is a non-trivial problem, which attracted considerable research attention in the past few years. In this talk, I will present the categorization of graph-based deep learning methods and review these methods following their history of developments and how these methods solve challenges of graphs. The differences of these models and how to composite different architectures will also be discussed. Finally, I will discuss potential future directions. More details can be found in our survey paper: https://arxiv.org/abs/1812.04202.
November 19, 2018
Oral Presentation, ICDM, 2018, Sentosa, Singapore
Network embedding has attracted considerable research attention recently. However, the existing methods are incapable of handling billion-scale networks, because they are computationally expensive and, at the same time, difficult to be accelerated by distributed computing schemes.
August 22, 2018
Oral Presentation, KDD, 2018, London, UK
Network embedding has received increasing research attention in recent years. The existing methods show that the high-order proximity plays a key role in capturing the underlying structure of the network. However, two fundamental problems in preserving the high-order proximity remain unsolved. First, all the existing methods can only preserve fixed-order proximities, despite that proximities of different orders are often desired for distinct networks and target applications. Second, given a certain order proximity, the existing methods cannot guarantee accuracy and efficiency simultaneously.
July 22, 2018
Talk, KDD Summer School, Chengdu, China
Network embedding has received increasing research attention in recent years. The existing methods show that the high-order proximity plays a key role in capturing the underlying structure of the network. However, two fundamental problems in preserving the high-order proximity remain unsolved. First, all the existing methods can only preserve fixed-order proximities, despite that proximities of different orders are often desired for distinct networks and target applications. Second, given a certain order proximity, the existing methods cannot guarantee accuracy and efficiency simultaneously.
June 29, 2018
Talk, Special Interesting Group on Network Analysis and Learning (SigNAL), Key Laboratory of Network Data Science and Technology, CAS, Beijing, China
Network embedding, which represents nodes in the network by low dimensional vectors, has attracted increasing research attention in the past few years. Despite remarkable progress, many key problems in network embedding remain unexplored, especially when handling large-scale networks. In this talk, I will present our recent works on large-scale network embedding, including how to preserve arbitrary-order proximity, how to handle dynamic networks and how to conduct distributed computing. Possible future directions will also be discussed.
August 17, 2016
Oral Presentation, KDD, 2016, San Francisco, CA, USA
Graph embedding algorithms embed a graph into a vector space where the structure and the inherent properties of the graph are preserved. The existing graph embedding methods cannot preserve the asymmetric transitivity well, which is a critical property of directed graphs. Asymmetric transitivity depicts the correlation among directed edges, that is, if there is a directed path from u to v, then there is likely a directed edge from u to v. Asymmetric transitivity can help in capturing structures of graphs and recovering from partially observed graphs. To tackle this challenge, we propose the idea of preserving asymmetric transitivity by approximating high-order proximity which are based on asymmetric transitivity. In particular, we develop a novel graph embedding algorithm, High-Order Proximity preserved Embedding (HOPE for short), which is scalable to preserve high-order proximities of large scale graphs and capable of capturing the asymmetric transitivity. More specifically, we first derive a general formulation that cover multiple popular highorder proximity measurements, then propose a scalable embedding algorithm to approximate the high-order proximity measurements based on their general formulation. Moreover, we provide a theoretical upper bound on the RMSE (Root Mean Squared Error) of the approximation. Our empirical experiments on a synthetic dataset and three realworld datasets demonstrate that HOPE can approximate the high-order proximities significantly better than the state-ofart algorithms and outperform the state-of-art algorithms in tasks of reconstruction, link prediction and vertex recommendation.
