More about HKUST
Visual Analytics for Explainable AI and Automated Machine Learning
Speaker: Professor Huamin QU Department of Computer Science and Engineering Hong Kong University of Science and Technology Title: "Visual Analytics for Explainable AI and Automated Machine Learning" Date: Monday, 30 Sept 2019 Time: 4:00pm - 5:00pm Venue: Lecture Theater F (near lift no. 25/26), HKUST Abstract: With the recent advances in machine learning, especially deep learning, we have witnessed increasing applications of AI in various domains. The past few years have seen a growing interest towards explainable AI (XAI) and automated machine learning (AutoML). At the VisLab of Hong Kong University of Science and Technology, we focus on human-centered approaches to make AI explainable, interactive, and trustworthy. In critical domains such as finance, security, and healthcare, explainability allows users to make more reliable decisions powered by the collaboration of machine- and human-intelligence. Interactive machine learning includes human inputs in the model training process to improve the performance of models. With transparency and intuitive feedback to users, it is easier to build trustworthy AI solutions. In this talk, I will first outline the research issues in visual analytics for explainable AI and automated machine learning. After that I will introduce some representative works from the VisLab including 1) DeepTracker, a visual analytics system to facilitate the exploration of the rich dynamics of CNN training processes and to identify the unusual patterns that are hidden behind the huge amount of training log; 2) RNNVis, a visual analytics tool for understanding and comparing recurrent neural networks (RNNs) for text-based applications; 3) RuleMatrix, an interactive visualization technique that helps users with little expertise in machine learning to understand, explore and validate classification models using rule-based explanations; 4) ATMSeer, an interactive visualization tool that supports machine learning experts in analyzing the results of AutoML and enables them to monitor the AutoML process, analyze the searched models, and refine the search space in real time. *************** Biography: Huamin Qu is a professor in the Department of Computer Science and Engineering (CSE) at the Hong Kong University of Science and Technology. His main research interests are in visualization and human-computer interaction, with focuses on urban informatics, social network analysis, e-learning, text visualization, and explainable artificial intelligence. His research has been recognized by 9 best paper/honorable mention awards, 2009 IBM Faculty Award, 2014 Higher Education Scientific and Technological Progress Award (Second Class) from the Ministry of Education of China, 2015 HKICT Best Innovation (Innovative Technology) Silver Award from the Hong Kong Institution of Engineers, 2015 APICTA Merit Award in E-Learning from the Asia Pacific ICT Alliance, 2016 Distinguished Collaborator Award from Huawei Noah's Ark Lab, and 2018 Yelp Dataset Challenge Round 10 Grand Prize Award. He is currently an associate editor of Computer Graphics Forum (CGF), and was an associate editor of IEEE TVCG, a paper co-chair for IEEE VIS'14, VIS'15 and VIS'18. He obtained a BS in Mathematics from Xi'an Jiaotong University, China, an MS and a PhD (2004) in Computer Science from the Stony Brook University.