讲座时间：2014年7月5日（周六），10:30 -11:30
讲座地点：西南交通大学九里校区交通运输与物流学院 01109
主讲人简介：成秀珍 教授，美国华盛顿大学
Xiuzhen Cheng received her MS and PhD degrees in computer science from the University of Minnesota -- Twin Cities, in 2000 and 2002, respectively. She is a professor at the Department of Computer Science, The George Washington University, Washington DC. Her current research interests focus on cognitive radio networks, wireless and mobile security, mobile handset networking systems (mobile health and safety), wireless and mobile computing, sensor networking, and algorithm design and analysis. She has served on the editorial boards of several technical journals (e.g., IEEE Transactions on Parallel and Distributed Systems, IEEE Wireless Communications, etc.) and the technical program committees of various professional conferences/workshops (e.g., IEEE INFOCOM, IEEE ICDCS, IEEE/ACM IWQoS, etc). She is the steering committee chair of the International Conference in Wireless Algorithms, Systems, and Applications (WASA). She worked as a program director for the US National Science Foundation (NSF) from April to October in 2006 (full time), and from April 2008 to May 2010 (part time). She is a senior member of IEEE and a member of ACM.
讲座内容简介：
题 目：Self-Collusion Resistant Auctions in Heterogeneous Spectrum Markets
Over the past few years a number of auction mechanisms have been proposed to improve the channel utilization while benefitting both the primary and the secondary users in a spectrum market. The major design objective of these auctions focuses on truthfulness to prevent market manipulation, i.e., to ensure that no buyer/seller can obtain a larger utility via cheating on its bid price. Nevertheless, self-collusion resulted from bid diversity in heterogeneous spectrum markets has been largely overlooked by the current research. Moreover, a systematic study to reveal the correlations among the three classic auctions (VCG, Myerson's Optimal Mechanism (MOM), and McAfee), based on which the current spectrum auctions are designed, is missing, making it hard to select an appropriate auction mechanism for practical settings. In this talk, we will discuss the challenges brought by the spectrum reuse and report our recent research in self-collusion resistant and truthful auction design. Particularly, we will present a collusion-resistant truthful auction framework, which can yield efficient auctions (like VCG), (sub)optimal auctions (like MOM), and attribute-aware auctions, by properly setting one simple parameter. Finally, we will discuss a few open research issues faced by collusion resistant auctions in more practical heterogeneous spectrum markets.