The 2010 Pacific-Rim Conference on Multimedia (PCM 2010)
September 21-24, Shanghai, China
About PCM
Organizing Committee
Technical Committee
Keynote Speakers
Conference Program
Info for Presenters
Venue and Hotel

Call for Papers
Call for Special Sessions
Paper Submission
Tutorial Session
Special Session
Author's Guide
Registration (new!)

About Shanghai
About World Expo
About China

Contact Us


Lecture Notes in Computer Science

Asia-Pacific Signal and Information Processing Association

Fudan University



Tie-Yan LIU

Dr. Tie-Yan LIU 

Title: Learning to Rank: Pushing the Frontier of Web Search


Abstract: Web Search has experienced an explosive growth in the past decades. In late 1990s, link analysis, which determines the importance of web pages, surges up the first wave of web search technologies and differentiates web search engines from conventional digital libraries. This directly leads to the huge success of Google-like companies. In recent years, learning to rank has emerged as the second wave of web search technologies, which enables newly-born search engines to catch up with those pioneers very quickly, even if they are lack of the accumulation of ranking heuristics through many years of operations. The increase of market share for Microsoft well demonstrates the power of learning to rank technologies.


In this talk, I will introduce the recent advances in learning to rank. In particular, I will focus on a newly developed approach, which we call listwise learning to rank. This approach differs from conventional machine learning methods in treating the ranked lists (permutations) of documents as learning instances, and defining loss functions according to the evaluation measures for search. I will introduce the representative listwise learning-to-rank algorithms, demonstrate their empirical effectiveness, and analyze their theoretical properties such as generalization ability and statistical consistency. Furthermore, I will make discussions on the future opportunities in learning to rank, for both academia and industry.


Bio: Tie-Yan Liu is a lead researcher at Microsoft Research Asia. His research interests include learning to rank, large-scale graph learning, and online advertising. So far, he has more than 70 quality papers published in referred conferences and journals and over 50 filed US / international patents or pending applications. He is the co-author of the Best Student Paper for SIGIR (2008), and the Most Cited Paper for the Journal of Visual Communication and Image Representation (2004~2006). He is a Program Committee Co-Chair of RIAO (2010), a Track Chair of WWW (2011), an Area Chair of SIGIR (2008~2010) and AIRS (2009, 2010), a Co-Chair of SIGIR workshop on learning to rank for IR (2007~2009), ICML workshop on learning to rank challenge (2010), NIPS workshop on machine learning in online advertising (2010), and a Program Committee member of many other international conferences. He is on the Editorial Board of the Information Retrieval Journal (IRJ), and is a guest editor of the special issues on learning to rank at IRJ and Journal of Machine Learning Research. He has given tutorials on learning to rank at several conferences including SIGIR (2008, 2010), WWW (2008, 2009), and AIRS (2008). Prior to joining Microsoft, he obtained his Ph.D. in electronic engineering from Tsinghua University. He is a senior member of the IEEE.




Yong RUI

Dr. Yong RUI

Title: The Evolution of Image Search


Abstract: Multimedia, especially image, search has emerged as an important research topic. For the past 40 years, there exist three paradigms on the research spectrum that ranges from the least automatic to the most automatic. On the far left end, there is the pure manual labeling paradigm that labels images manually with text labels and then use text search to search images indirectly. On the far right end, there is the content-based search paradigm that can be fully automatic by using low-level features from image content analysis. In recent years, a third paradigm emerged which is in the middle: the annotation-learning paradigm. Once the concept models are trained (by professional labeled data (1.0) or user labeled data (2.0)), this paradigm can automatically understand concepts in unseen images. This talk looks into this annotation-learning paradigm. Specifically, this talk explores current (e.g., relationship based) and future (user labeled) directions on this paradigm.


Bio: Yong Rui is a Senior Director of Microsoft Asia R&D Group responsible for Technology and Product Innovation. From 2006 to 2008, he was a Director responsible for R&D strategy in China; from 2008 to 2010, he was a Director responsible for Microsoft Education Product; and from 1999 to 2006, he was leading the Multimedia Collaboration team at Microsoft Research, Redmond, USA.


In end of 2009, Dr. Rui became one of the 7 Fellows of IEEE worldwide who are younger than 40 years old.  He is a member of National Academy of Engineering’s Symposium of Frontiers of Engineering.  He serves as a member of review panels for both the US National Science Foundation (NSF) and the Chinese National Science Foundation.  He is a member of review committee of China’s National 1000-People Program.


Dr. Rui is recognized as a pioneer in multimedia search. He published sixteen books and book chapters, and 100+ referred journal and conference papers. His publications are among the most cited. According to Google Scholar citation, two of Dr. Rui’s papers have been each cited 1,000+ times and 12 additional papers have each been cited 100+ times.  


Dr. Rui is the Associate Editor-in-Chief of IEEE Multimedia Magazine, an Associate Editor of ACM Trans. on Multimedia Computing, Communication and Applications (TOMCCAP), and an Associate Editor of IEEE Trans. on Circuits and Systems for Video Technologies (CSVT). He was an Associate Editor of IEEE Trans. on Multimedia (2004-2008), ACM/Springer Multimedia Systems Journal (2004-2006), and International Journal of Multimedia Tools and Applications (2004-2006). He also serves on the Advisory Board of IEEE Trans. on Automation Science and Engineering.  Dr. Rui is on Organizing Committees and Program Committees of numerous conferences including ACM Multimedia, IEEE CVPR, IEEE ECCV, IEEE ACCV, IEEE ICIP, IEEE ICASSP, IEEE ICME, SPIE ITCom, ICPR, CIVR, among others. He was Program Co-Chair of ACM Multimedia 2006, Program Co-Chair of Pacific Rim Multimedia (PCM) 2006, General Co-Chair of Conference on Image and Video Retrieval (CIVR) 2006, Program Co-Chair of IEEE ICME 2009 and General Co-Chair of ACM Multimedia 2009.  Dr. Rui is on the ACM SIG Multimedia Executive Committee, and Steering Committees of IEEE ICME, ACM ICMR and PCM.   He is the Chair of ACM SIG Multimedia Beijing Chapter.  


Dr. Rui is a Fellow of IEEE and a Distinguished Member of ACM.




Zhi-Hua ZHOU

Prof. Zhi-Hua ZHOU

Title: A New Machine Learning Framework with Application to Image Annotation


Abstract: Machine learning has been recognized as a powerful technique to help improving performance in various multimedia applications. Generally, each real object is represented by a feature vector (or instance) associated with a class label indicating the semantic meaning of that object. For ambiguous objects which have multiple semantic meanings, such as images in multimedia applications, traditional machine learning frameworks may be less effective. This talk will introduce the MIML (Multi-Instance Multi-Label learning) framework which provides a promising direction for learning with such objects, and its application to a real-world image annotation task.


Bio: Zhi-Hua Zhou is a professor at the Department of Computer Science and Technology, Nanjing University, China. His research interests are mainly in machine learning, data mining, pattern recognition and image retrieval. In these areas he has published over 80 papers in leading international journals or conferences, and holds 11 patents. He is an Associate Editor-in-Chief of "Chinese Science Bulletin", Associate Editor of "IEEE Transactions on Knowledge and Data Engineering" and "ACM Transactions on Intelligent Systems and Technology", and on the editorial boards of various other journals. He also serves/served as guest editor for various journals including “Machine Learning”, “IEEE Intelligent Systems”, “Pattern Recognition”, " Multimedia Systems", etc. He is the Founding Steering Committee Co-Chair of ACML, and Steering Committee member of PAKDD and PRICAI. He served as Program Committee Chair/Co-Chair of PAKDD'07, PRICAI'08 and ACML'09, Vice Chair or Area Chair or Senior Program Committee member of many conferences such as KDD’10, ECML PKDD’10, ICPR’10, SDM’09, CIKM’09, ICDM’08, etc. He is the Chair of the Machine Learning Society of the China Association of Artificial Intelligence (CAAI), the Vice Chair of the Artificial Intelligence and Pattern Recognition Society of the China Computer Federation (CCF), and the Chair of the IEEE Computer Society Nanjing Chapter.