Prof. Jian Pei
Simon Fraser University, Canada
Jian Pei is a Professor at Simon Fraser University. His research focuses on data science, big data, data mining, database systems, and information retrieval. His expertise is in developing effective and efficient data analysis techniques for novel data intensive applications and transferring his research results to industry products and business practice. He is recognized as a Fellow of the Royal Society of Canada (Canada's national academy), the Canadian Academy of Engineering, ACM, and IEEE. Since 2000, he has published one textbook, two monographs and over 300 research papers in refereed journals and conferences, which have been cited extensively. He was the editor-in-chief of the IEEE Transactions of Knowledge and Data Engineering (TKDE) in 2013-16, the chair of ACM SIGKDD in 2017-2021, and the organizers of many conferences. He received a few prestigious awards, including the 2017 ACM SIGKDD Innovation Award, the 2015 ACM SIGKDD Service Award, the 2014 IEEE ICDM Research Contributions Award, the British Columbia Innovation Council 2005 Young Innovator Award, an IBM Faculty Award, a KDD Best Application Paper Award, and an ICDE Influential Paper Award.
Speech Title: Exact, Concise, and Consistent Data Driven Interpretation
Abstract: Interpretability and explainability are at the core in human being's pursuit of new knowledge. At the same time, interpretation in data analytics and data mining is challenging in many ways, such as the complexity of models to be interpreted, the difficulty in knowledge elicitation, the expectation of embodying interpretation, and the need of many kinds of knowledge. In this talk, I will present our systematic research on exact, concise, and consistent data driven interpretation for database and data mining tasks. I will illustrate our principles and techniques using several application examples, including multidimensional skyline queries (aka pareto optima) in databases, piece-wise linear neural networks in classification, and KS-tests in statistics. I will also discuss the promises and challenges of data driven interpretation for future work.
Prof. Ji-Rong Wen
Renmin University of China, China
Ji-Rong Wen is a full professor, the dean of School of Information, and the executive dean of Gaoling School of Artificial Intelligence at Renmin University of China (RUC). He has been working in the big data and AI areas for many years. He is the PC Chair of SIGIR 2020 and the Associate Editor of ACM TOIS and IEEE TKDE. He worked at Microsoft Research Asia (MSRA) for 14 years and once was a senior researcher and the group manager of the Web Search and Mining Group. He was elected as a National Distinguished Professor in 2013 and Beijing’s Distinguished Young Scientist in 2018. He is a Chief Scientist of Beijing Academy of Artificial Intelligence.
Speech Title: Wenlan - A Large-scale Multi-modal Pre-trained Model
Abstract: In this talk, I will introduce our recent work on a large-scale multi-modal pre-trained model named Wenlan. Wenlan was trained in a self-supervised way on huge datasets containing billions of image-text pairs collected from the internet and millions of short videos. We have applied Wenlan to a number of downstream tasks and demonstrated its superiority and versatility. Moreover, I will share some insights by further exploring and exploiting the Wenlan model, which verify that multi-modal pre-training is a promising way to get better representation, structure and knowledge like human beings.
Prof. Bernhard Schölkopf
Max Planck Institute for Intelligent Systems, Germany
Bernhard Schölkopf's scientific interests are in machine learning and causal inference. He has applied his methods to a number of different fields, ranging from biomedical problems to computational photography and astronomy. Bernhard has researched at AT&T Bell Labs, at GMD FIRST, Berlin, and at Microsoft Research Cambridge, UK, before becoming a Max Planck director in 2001. He is a member of the German Academy of Sciences (Leopoldina), has (co-)received the Royal Society Milner Award, the Leibniz Award, the Koerber European Science Prize, and the BBVA Foundation Frontiers of Knowledge Award, He is Fellow of the ACM and of the CIFAR Program "Learning in Machines and Brains", an Amazon Distinguished Scholar, and holds a Professorship at ETH Zurich. Bernhard co-founded the series of Machine Learning Summer Schools, and helped build the Journal of Machine Learning Research, an early development in open access and today the field's flagship journal.
Speech Title: From Statistical to Causal Machine Learning
Abstract: In machine learning, we use data to automatically find dependences in the world, with the goal of predicting future observations. Its methods build on statistical dependences, but one can try to go beyond this, assaying underlying causal structures. Causal models may be more robust to changes that occur in real world datasets and thus play a central role in addressing some of the hard open problems of the field. We discuss implications of causal models for machine learning, as well as connections in the opposite direction, including the prospects of causal representation learning.