报告时间：2016年5月6日（星期五）10:00-12:00
报告地点：浙江大学紫金港校区蒙民伟楼CAD&CG国家重点实验室402室
报告人：Prof. Hui Fang Prof. Xiaoming Li
主持人：蔡登 教授
报告一: Opinions matter: a general approach to user profile modeling for contextual suggestion
Abstract: The increasing use of mobile devices enables an information retrieval (IR) system to capitalize on various types of contexts (e.g., temporal and geographical infor- mation) about its users. Combined with the user preference history recorded in the system, a better understanding of users’ information need can be achieved and it thus leads to improved user satisfaction. More importantly, such a system could proactively recommend suggestions based on the contexts. User profiling is essential in contextual suggestion. However, given most users’ observed behaviors are sparse and their preferences are latent in an IR system, constructing accurate user profiles is generally difficult. In this task, I will present our recent work on location-based contextual suggestion. In particular, we propose to leverage users’ opinions to construct the profiles. Instead of simply recording ‘‘what places a user likes or dislikes’’ in the past (i.e., description-based profile), we want to construct a profile to identify ‘‘why a user likes or dislikes a place’’ so as to better predict whether the user would like a new candidate suggestion of place. Candidate suggestions are represented in the same fashion and ranked based on their similarities with respect to the user profiles. Moreover, we also develop a novel summary generation method that utilizes the opinion-based user profiles to generate personalized and high-quality summaries for the suggestions. The systems developed based on the proposed methods have been ranked as top 1 in both TREC 2013 and 2014 contextual suggestion tracks.
Bio: Hui Fang is an Associate Professor in the Department of Electrical and Computer Engineering at University Delaware. She received her M.S. and Ph.D. degree from University of Illinois at Urbana-Champaign in 2004 and 2007, respectively, and B.S degree from Tsinghua University in 2001. Her primary research interest is information retrieval, with focus on enterprise search, search results diversification and axiomatic retrieval models. She received the ACM SIGIR 2004 Best Paper Award and 2010-2011 HP Labs Innovation Research Awards.
报告二: High-performance Computing on GPU: Introduction and Challenges
Abstract: Graphic Processing Units (GPU) and the many-simple-core processor family present a promising direction for conducting traditionally elite-only high-performance computing on commodity hardware. In this talk, I will discuss why we use GPU for computing, what are the main technical challenges if you want to adopt GPU for your applications, and how real-world large scale applications benefit from the new processor. In particular, I will demonstrate the techniques developed in our group to tackle some difficult and basic software problems in the development GPU-oriented software, and showcase our unique integration of mathematical transformations and program optimization that lead to one of the fastest FFT libraries on GPU.
Bio: Xiaoming Li obtained his B.S. and M.E. in Computer Science at Nanjing University in 1998 and 2001, and Ph.D. in Computer Science at the University of Illinois at Urbana-Champaign in 2006. He is currently an Associate Professor of Computer Engineering at the University of Delaware. His research interests are high-performance computing fundamentals and applications, in particular on code generation and optimization, compilers, and interaction between hardware and software. The main goal of his research is to make programs run faster and use less resource. His research covers all aspect of compiler optimizations and transformations, as well as domain-specific tuning for computation-driven applications.
