BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Rencang Li (University Of Texas At Arlington) DTSTART:20201228T013000Z DTEND:20201228T023000Z DTSTAMP:20260423T024658Z UID:iccm2020/32 DESCRIPTION:Title: Orthogonal Multi-View Subspace Learning\nby Rencang Li (University O f Texas At Arlington) as part of ICCM 2020\n\n\nAbstract\nMulti-view data are increasingly collected for a variety of applications in the real world . They contain complementary\, redundant\, and corroborative contents and so provide more informative than single_x005fview data when it comes to ch aracterize objects of the real-world. It is rather natural for human being s to perceive the world through comprehensive information collected by mul tiple sensory organs\, but it is an open question on how to endow machines with analogous cognitive capabilities to do the same.One of the fundament al challenges is how to represent and summarize multi-view data in such a way that comprehensive information concealed in multi-view data can be pro perly exploited by multi-view learning models. In this talk\, we will pres ent a unified framework for multi-view subspace learning to learn individu al orthogonal projections for all views. The framework integrates the corr elations within multiple views\, supervised discriminant capacity\, and di stance preservation in a concise and compact way. It not only includes sev eral existing models as special cases\, but also inspires new novel models . Besides the framework\, we will discuss highly efficient numerical metho ds to solve the associated optimization problems. The methods are built up on an iterative Krylov subspace method which can easily scale up for large size datasets. Extensive experiments are conducted on various real-world datasets for the multi-view discriminant analysis and multi-view multi-lab el classification tasks. The experimental results demonstrate that the pro posed models are consistently competitive to and often better than the sta te-of-the-art methods.This is a joint work with Li Wang (UT Arlington)\, L ei-hong Zhang (Soochow University)\, and Chungen Shen (University of Shang hai for Science and Technology).\n LOCATION:https://researchseminars.org/talk/iccm2020/32/ END:VEVENT END:VCALENDAR