BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Memento EPFL// BEGIN:VEVENT SUMMARY:AI Center Seminar - AI Fundamentals series - Prof. Selin Aviyente DTSTART:20260928T110000 DTEND:20260928T120000 DTSTAMP:20260918T053051Z UID:42c292747b9613931b84c2de9637f43e7d7bf9b3da798aca4750a400 CATEGORIES:Conferences - Seminars DESCRIPTION:Prof. Selin Aviyente\nThe talk is jointly organized by the  EPFL AI Center and the EPFL Signal Processing Laboratory (LTS4) as part of the AI Fundamentals seminar series.\n\nHost: Dr. Dorina Thanou\n\nTitl e \nMultiview Graph Learning: Algorithms and Applications\n\nAbstract\nIn many modern data science applications\, relationships between data sample s are well described with a graph structure. While many real-world data ar e intrinsically graph-structured\, there are a large number of application s\, particularly in biology and neuroscience\, where the graph is not read ily available and needs to be learned from a set of observations\, i.e. gr aph signals. Most of the existing work on graph learning assumes homogeneo us data defined on a single\, undirected graph. However\, in many settings \, the data are heterogeneous and arise from multiple related graphs\, ref erred to as a multiview graph. For example\, neuroimaging data across mult iple subjects can be modeled as a multiview graph in which each view corre sponds to an individual brain connectome. In these settings\, the views of the multiview graph are closely related to each other. Therefore\, learni ng the topology of views jointly by incorporating the relationships among views can substantially improve performance over learning each view indepe ndently. \n\nIn this talk\, I introduce a framework for multiview graph l earning (mvGL) built on two complementary models. In the first\, the views are assumed to be generated from a shared consensus graph through a pertu rbation function\, capturing the common structure underlying all views. In the second\, structural similarity across views is driven by the connecti ons of common hub nodes\, enabling a node-level characterization of cross- view relationships. I present results on both simulated data and real neur oimaging datasets\, demonstrating the effectiveness of the proposed method s in recovering meaningful graph structure from multiview observations.\n\ n\nBio\nSelin Aviyente received her B.S. degree in Electrical and Electron ics engineering from Bogazici University\, Istanbul in 1997\; M.S. and Ph. D. degrees\, both in Electrical Engineering: Systems\, from the University of Michigan\, Ann Arbor\, in 1999 and 2002\, respectively. She joined the Department of Electrical and Computer Engineering at Michigan State Unive rsity in 2002\, where she is currently James O. Fishbeck and Lee A. Morgan Professor. Her research focuses on statistical and nonstationary signal p rocessing\, higher-order data representations and network science with app lications to biological signals. She has authored more than 150 peer-revie wed journal and conference papers. She is the recipient of a 2005 Withrow Teaching Excellence Award\, a 2008 NSF CAREER Award and 2021 Withrow Excel lence in Diversity Award. She has served as the chair of IEEE Signal Proce ssing Society Bioimaging and Signal Processing Technical Committee and is currently serving on IEEE Signal Processing Society Machine Learning for S ignal Processing Technical Committee\, Steering Committees of IEEE SPS Dat a Science Initiative and IEEE BRAIN. She is the Area Editor for Special Is sues for IEEE Signal Processing Magazine and Senior Area Editor for IEEE T ransactions on Signal and Information Processing over Networks. She is a 2 025 IEEE Signal Processing Society Distinguished Lecturer.\n  LOCATION:ELE 117 https://plan.epfl.ch/?room==ELE%20117 https://epfl.zoom.u s/j/63145150854?pwd=0fXeWpshf2a5fHl9bvjtRrpN7blhnP.1 STATUS:CONFIRMED END:VEVENT END:VCALENDAR