"Big Data: An Imbalanced Learning Perspective"
Professor Haibo He
Editor-in-Chief, IEEE Transactions on Neural Networks and Learning Systems
Robert Haas Endowed Chair Professor,
Department of Electrical, Computer, and Biomedical Engineering,
University of Rhode Island, USA
Big data has become an important topic worldwide over the past several years. Among many aspects of the big data research and development, imbalanced learning has become a critical component as many data sets in real-world applications are imbalanced, ranging from surveillance, security, Internet, finance, social network, to medical and healthy related data analysis. In general, the imbalanced learning problem is concerned with the performance of learning algorithms in the presence of underrepresented data and severe class distribution skews. Due to the inherent complex characteristics of imbalanced data sets, learning from such data requires new understandings, principles, algorithms, and tools to transform vast amounts of raw data efficiently and effectively into information and knowledge representation.
In this talk, I will start with an overview of the nature and foundation of the imbalanced learning, and then focus on the state-of-the-art methods and technologies in dealing with the imbalanced data, followed by a systematic discussion on the assessment metrics to evaluate learning performance under the imbalanced learning scenario. I will also introduce the latest imbalanced learning methods that we developed to handle different types of imbalanced data sets. Finally, I will highlight the major opportunities and challenges, as well as potential important research directions for learning from imbalanced data facing the big data era.
Haibo He is a Fellow of IEEE and the Robert Haas Endowed Chair Professor at the University of Rhode Island, Kingston, RI, USA. His primary research interests include computational intelligence and various applications. He has published one sole-author book (Wiley), edited 1 book (Wiley-IEEE) and 6 conference proceedings (Springer), and authored/co-authors over 280 peer-reviewed journal and conference papers, including several highly cited papers in IEEE Transactions on Neural Networks and IEEE Transactions on Knowledge and Data Engineering, Cover Page Highlighted paper in IEEE Transactions on Information Forensics and Security, and Best Readings of the IEEE Communications Society. He has delivered more than 50 invited talks around the globe. He was the Chair of IEEE Computational Intelligence Society (CIS) Emergent Technologies Technical Committee (ETTC) (2015) and the Chair of IEEE CIS Neural Networks Technical Committee (NNTC) (2013 and 2014). He served as the General Chair of 2014 IEEE Symposium Series on Computational Intelligence (IEEE SSCI’14, Orlando, Florida). He is currently the Editor-in-Chief of IEEE Transactions on Neural Networks and Learning Systems. He was a recipient of the IEEE International Conference on Communications (ICC) “Best Paper Award” (2014), IEEE CIS “Outstanding Early Career Award” (2014), National Science Foundation “Faculty Early Career Development (CAREER) Award” (2011), and Providence Business News (PBN) “Rising Star Innovator” Award (2011). More information can be found at: http://www.ele.uri.edu/faculty/he/
"Data-driven Surrogate-assisted Evolutionary Optimization of Expensive Optimization Problems"
Professor Yaochu Jin
Editor-in-Chief, IEEE Transactions on Cognitive and Developmental Systems
Editor-in-Chief, Complex & Intelligent Systems
Professor in Computational Intelligence,
Head of the Nature Inspired Computing and Engineering (NICE) group,
Co-Coordinator of the Centre for Mathematical and Computational Biology (CMCB),
Department of Computer Science,
University of Surrey, UK
This talk discusses the main challenges in data-driven surrogate-assisted evolutionary optimization of expensive problems. Fundamental issues such as surrogate model selection, surrogate model management and model training using advanced machine learning techniques in single and multi- and many-objective optimization will be discussed. Challenges in handing sparse data or big data and recent advances in surrogate-assisted optimization of high-dimensional expensive optimization problems will be presented. Finally, a few real-world examples for data-driven optimization and decision-making will be given.
Yaochu Jin is a Professor in Computational Intelligence, Head of the Nature Inspired Computing and Engineering (NICE) group, Co-Coordinator of the Centre for Mathematical and Computational Biology (CMCB), Department of Computer Science, University of Surrey. He is also a Finland Distinguished Professor (2015-17) with the Industrial Optimization Group, Department of Mathematical Information, University of Jyvaskyla, Finland, and a Changjiang Distinguished Visiting Professor (2015-17), State Key Laboratory of Synthetical Automation of Process Industry, Northeastern University, China.
Professor Jin is the Editor-in-Chief of the IEEE Transactions on Cognitive and Developmental Systems and Co-Editor-in-Chief of Complex & Intelligent Systems (Springer). He was an elected AdCom member (2012-2013), Vice President for Technical Activities (2014-2015), and an IEEE Distinguished Lecturer (2013-2015, 2017-2019) of the IEEE Computational Intelligence Society. He was elevated to an IEEE Fellow for contributions to evolutionary optimization.
Professor Jie Lu
IEEE Fellow, IFSA Fellow
Editor-In-Chief, Knowledge-Based Systems
Editor-In-Chief, International Journal of Computational Intelligence Systems
Director of Centre for Artificial Intelligence
Associate Dean (Research Excellence) and Distinguished Professor,
Faculty of Engineering and IT,
University of Technology Sydney, Australia
Concept Drift is known as unforeseeable change in underlying streaming data distribution over time. The phenomenon of concept drift has been recognized as the root cause of decreased effectiveness in many decision-related applications. Adaptive learning under concept drift is a relatively new research field and is one of the most pressing and fundamental problems in the current age of big data. Building an adaptive system is a highly promising solution for coping with persistent environmental change and avoiding system performance degradation. This talk will present a set of methods and algorithms that can effectively and accurately detect concept drift, understand and react to it, with knowledge adaptation, in a timely way.
Distinguished Professor Jie Lu is an internationally renowned scientist in the areas of computational intelligence, specifically in decision support systems, fuzzy transfer learning, concept drift, and recommender systems. She is the Associate Dean in Research Excellence in the Faculty of Engineering and Information Technology at University of Technology Sydney (UTS) and the Director of Centre for Artificial Intelligence (CAI) at UTS. She is also the co-Director of the Joint Research Centre Wise Information Systems (WIS) between UTS and Shanghai University. She has published six research books and 400 papers in Artificial Intelligence, IEEE transactions on Fuzzy Systems and other refereed journals and conference proceedings (H-index 44, Google Scholar). She has won eight Australian Research Council (ARC) discovery grants and 10 other research grants for over $4 million. She serves as Editor-In-Chief for Knowledge-Based Systems (Elsevier) and Editor-In-Chief for International Journal on Computational Intelligence Systems (Atlantis), has delivered 20 keynote speeches at international conferences, and has chaired 10 international conferences. She is a Fellow of IEEE and Fellow of IFSA.
"Reproducibility in Big Data Analysis: A Bad Data Perspective"
Professor Zidong Wang
Professor of Dynamical Systems and Computing,
Department of Computer Science,
Brunel University London, UK
In this talk, we discuss another side of big data analysis, bad data analysis, where the badness means the complexities resulting in the reproducibility issues. Some background knowledge is first introduced on the volatility of the big data analysis, which shows 1) “big” does not necessarily mean “better” and 2) the so-called multi-objective data analysis (against badness) is vitally important in advancing the state-of-the-art. Two examples are used for demonstration of the big data analysis, one for big data from complex networks and the other for big data from gene expression image processing. Finally, conclusions are drawn and some future directions are pointed out.
Zidong Wang is an IEEE Fellow and Professor of Computing at Brunel University London with research interests in intelligent data analysis, statistical signal processing as well as dynamic systems and control. He has been named as the Hottest Scientific Researcher in 2012 in the area of Big Data Analysis (see http://sciencewatch.com/articles/hottest-research-2012). He was awarded the AvH Research Fellowship in 1996 from the Alexander von Humboldt Foundation of Germany, the JSPS Research Fellowship in 1998 from the Japan Society for the Promotion of Science and the William Mong Distinguished Research Fellowship in 2002 from the University of Hong Kong. Since 1997, He has published around 310 papers in prestigious international journals (including 110 papers in IEEE Transactions) with h-index 60 according to the Web of Science. He is currently serving as an Associate Editor for 12 prestigious journals including 5 IEEE Transactions. His research has been funded by the EU, the Royal Society and the EPSRC.