Automated Learning form Graph-Structured Data


Graph-structured data (GSD) is ubiquitous in real-life applications, which appears in many learning applications such as property prediction for molecular graphs, product recommendations from heterogeneous information networks, and logical queries from knowledge graphs. Recently, learning from graph-structured data has also become a research focus in the machine learning community. However, again due to such diversities in GSD, there are no universal learning models that can perform well and consistently across different learning applications based on graphs. In sharp contrast to this, convolutional neural networks work well on natural images, and transformers are good choices for text data. In this tutorial, we will talk about using automated machine learning (AutoML) as a tool to design learning models for GSD. Specifically, we will elaborate on what is AutoML, what kind of prior information from graphs can be explored by AutoML, and how can insights be generated from the searched models.


Dr. Quanming Yao
Dr. Quanming Yao is a tenure-track assistant professor in the Department of Electronic Engineering, Tsinghua University. He was a senior scientist in 4Paradigm Inc., who is also the founding leader of the company’s machine learning research team. He obtained his Ph.D. degree at the Department of Computer Science and Engineering of Hong Kong University of Science and Technology (HKUST). His research interests are in machine learning, optimization, and automated machine learning. He has 40+ top-tier journal and conference papers, including ICML, NeurIPS, JMLR, and TPAMI, with a citation of 2300 (since 2015). He is a receipt of Forbes 30 Under 30 (China), Young Scientist Awards (issued by Hong Kong Institution of Science), Wuwen Jun Prize for Excellence Youth of Artificial Intelligence (issued by CAAI), and a winner of Google Fellowship (in machine learning).

Dr. Huan Zhao
Dr. Huan Zhao is a senior scientist in 4Paradigm Inc., China, leading the research on automated graph representation learning (AutoGraph) in the company. Prior to 4Paradigm, He worked as a research intern and senior algorithm engineer in Alibaba from November 2017 to July 2019. He obtained his Doctor Degree at the Department of Computer Science and Engineering, HKUST in Jan. 2019. His research interest includes recommender system, graph representation learning, and automated machine learning. He has published more than 20 top-tier conference and journal papers, including KDD, CIKM, AAAI, TKDE, TKDD, with a citation of 782 in Google Scholar (since 2016). Besides research, he also lead a team in 4Paradigm to deliver AutoGraph algorithms to real-world applications, retailing recommendation, financial fraud detection, bioinformatics, etc., from the business partners of 4Paradigm.

Dr. Yongqi Zhang
Dr. Yongqi Zhang is a research scientist in 4Paradigm. He obtained his Ph.D. degree at the Department of Computer Science and Engineering of Hong Kong University of Science and Technology (HKUST) in 2020 and received his bachelor degree at Shanghai Jiao Tong University (SJTU) in 2015. He has published four top-tier conference/journal papers as first-author in NeurIPS, ICDE, VLDB-Journal. His research interests focus on knowledge graph embedding, automated machine learning and deep learning. He was a Program Committee for AAAI 2020-2021, IJCAI 2020-2021, CIKM 2021, and a reviewer for TKDE and NETNET journals.