Dear Colleagues, I am a Ph.D student from Tsinghua University, Beijing, China. My research topic relates to applying graphical models to solve computer vision problems. In many cases, we need to inferring the MAP label configuration for a graphical model with cycles. Therefore, we have to turn to some approximate algorithms. However, there seems to existing so many choices. Many different approximate algorithms have been proposed. My question is, whether there exists some evaluation showing the pros and cons of each approach? Anyone can point me a link some papers or such work? Thanks very much.
Regards, Jinhui [EMAIL PROTECTED]
_______________________________________________ uai mailing list uai@ENGR.ORST.EDU https://secure.engr.oregonstate.edu/mailman/listinfo/uai
