站内搜索:
当前位置: 首页 >> 新闻中心 >> 通知公告 >> 正文

【信息学院学术报告】AnInformation-Theoretic Approach for Graphs

[发表时间]:2018-08-17 [来源]: [浏览次数]:

报告主题:AnInformation-Theoretic Approach for Graphs.

报告时间:2017年8月21日下午10:15—11:00

报告地点:沙河校区楼7号学院楼214

报告简介:Theestimation of mutual information between graphs has been an elusive problemuntil the formulation of graph matching in terms of manifold alignment. Then,graphs are mapped to multi-dimensional sets of points through structurepreserving embeddings. Point-wise alignment algorithms can be exploited in thiscontext to re-cast graph matching in terms of point matching. Methods based onbypass entropy estimation must be deployed to render the estimation of mutualinformation computationally tractable. In this paper the novel contribution isto show how manifold alignment can be combined with copula-based entropyestimators to efficiently estimate the mutual information between graphs. Wecompare the empirical copula with an Archimedean copula (the independent one)in terms of retrieval/recall after graph comparison. Our experiments show thatmutual information built in both choices improves significantly state-of-theart divergences.

主讲人简介:Prof.Francisco Escolano received his Bachelors degree in Computer Science from thePolytechnical University of Valencia (Spain) in 1992 and his Ph degree inComputer Science from the University of Alicante in 1997. Since 1998, he is anAssociate Professor with the Department of Computer Science and ArtificialIntelligence of the University of Alicante. He has been post-doctoral fellowwith Dr. Norberto M. Grzywacz at the Biomedical Engineering Department of theUniversity of South California in Los Angeles, and he has also collaboratedwith Dr. Alan L. Yuille at the Smith-Kettlewell Eye Research Institute of SanFrancisco. Recently, he visited the Liisa Holm's Bioinformatics Lab at theUniversity of Helsinki. His research interests are focused on the developmentof efficient and reliable computer-vision algorithms for biomedicalapplications (tracking of intravascular sequences), active vision and robotics(mid-level geometric structures obtained through junction grouping, stereo andappearance based methods for the localization of mobile robots, SLAM), andvideo-based surveillance (motion detection and object tracking). He is alsointerested in the coupling between computer and biological vision. He is thehead of the Robot Vision Group.

版权所有: 中央财经大学  学院南路校区地址:北京市海淀区学院南路39号 邮编:100081 沙河校区地址:北京市昌平区沙河高教园区 邮编:102206 京ICP备05004636号 京公网安备110402430071号