BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20250309T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20241103T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20250715T211650Z UID:BEC1A0FF-F1E4-4544-9C63-282349951E86 DTSTART;TZID=America/New_York:20250214T110000 DTEND;TZID=America/New_York:20250214T120000 DESCRIPTION:Euclidean mirrors and first-order changepoints in network time series\n\nDr. Zachary Lubberts\nAssistant Professor of Statistics\nUnivers ity of Virginia\n  \nFriday\, February 14th\, 2025 \n11:00 A.M. – 12:00 P.M. Eastern Time \nGeorge Mason University\nNguyen Engineering Bu ilding\, Jajodia Auditorium\, Room 1101 \n4511 Patriot Circle\, Fairfax\ , Virginia 22030\n\nThe seminar talk is also live-streamed. Please [regist er](https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fform s.office.com%2Fr%2FiKmfG9TRfL&data=05%7C02%7Ckhassan1%40gmu.edu%7Ce68f0039 cf52432c66c808dd4ab57147%7C9e857255df574c47a0c00546460380cb%7C0%7C0%7C6387 48865415063328%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjA uMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sda ta=1D2gQLkg7uv19bP2BlNWzEai3uqtgwiCO3DLlEQitbQ%3D&reserved=0) to receive t he link.\n \nAbstract: \nWe describe a model for a network time series whose evolution is governed by an underlying stochastic process\, known a s the latent position process\, in which network evolution can be represen ted in Euclidean space by a curve\, called the Euclidean mirror. We define the notion of a first-order changepoint for a time series of networks\, a nd construct a family of latent position process networks with underlying first-order changepoints. We prove that a spectral estimate of the associa ted Euclidean mirror localizes these changepoints\, even when the graph di stribution evolves continuously\, but at a rate that changes. Simulated an d real data examples on organoid networks show that this localization capt ures empirically significant shifts in network evolution.\n\nBio: \nZach Lubberts is a data scientist working on the interplay of statistics and o ptimization\, with an emphasis on statistics on graphs. Some of his recent publications concern accurate estimation of the eigenvectors of random ma trices and the capture of relevant signal in various graph models\, includ ing time series of graphs.\nLubberts earned his Ph.D. in applied mathemati cs and statistics from Johns Hopkins University in 2019. His dissertation research focused on the application of real algebraic geometry to the cons truction of multivariable tight wavelet frames for use in signal processin g. He also earned his bachelor’s degree in applied mathematics and stati stics and philosophy from Johns Hopkins in 2013.\n\nRoom: Jajodia Auditori um Room 1101\, Bldg: Nguyen Engineering Building\, \, George Mason Univers ity\, 4511 Patriot Circle\, Fairfax\, Virginia\, United States\, 22030\, V irtual: https://events.vtools.ieee.org/m/468380 LOCATION:Room: Jajodia Auditorium Room 1101\, Bldg: Nguyen Engineering Buil ding\, \, George Mason University\, 4511 Patriot Circle\, Fairfax\, Virgin ia\, United States\, 22030\, Virtual: https://events.vtools.ieee.org/m/468 380 ORGANIZER:kafi@ieee.org SEQUENCE:25 SUMMARY:Euclidean mirrors and first-order changepoints in network time seri es URL;VALUE=URI:https://events.vtools.ieee.org/m/468380 X-ALT-DESC:Description: <br /><div class="x_elementToProof"><strong><em dat a-olk-copy-source="MessageBody">Euclidean mirrors and first-order changepo ints in network time series</em></strong>&nbsp\;</div>\n<div aria-hidden=" true">&nbsp\;</div>\n<div><strong>Dr. Zachary Lubberts&nbsp\;&nbsp\;</stro ng>&nbsp\;</div>\n<div><strong>Assistant Professor of Statistics</strong>& nbsp\;</div>\n<div><strong>University of Virginia</strong>&nbsp\;</div>\n< div>  &nbsp\;</div>\n<div><strong>Friday\, February 14th\, 2025</stron g> &nbsp\;</div>\n<div><strong>11:00 A.M. &ndash\; 12:00 P.M. Eastern Ti me</strong> &nbsp\;</div>\n<div><strong>George Mason University</strong> </div>\n<div><strong>Nguyen Engineering Building\, Jajodia Auditorium\, Ro om 1101</strong> &nbsp\;</div>\n<div class="x_elementToProof"><strong>45 11 Patriot Circle\, Fairfax\, Virginia 22030</strong>&nbsp\;</div>\n<div a ria-hidden="true">&nbsp\;</div>\n<div class="x_elementToProof"><strong>The seminar talk is also live-streamed. Please <a id="x_anchor-99eec5d8-7979- 1ef8-6ed1-6413a1557c4e" class="x_OWAAutoLink" title="Original URL: https:/ /forms.office.com/r/iKmfG9TRfL. Click or tap if you trust this link." href ="https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fforms. office.com%2Fr%2FiKmfG9TRfL&amp\;data=05%7C02%7Ckhassan1%40gmu.edu%7Ce68f0 039cf52432c66c808dd4ab57147%7C9e857255df574c47a0c00546460380cb%7C0%7C0%7C6 38748865415063328%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIw LjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C& amp\;sdata=1D2gQLkg7uv19bP2BlNWzEai3uqtgwiCO3DLlEQitbQ%3D&amp\;reserved=0" target="_blank" rel="noopener noreferrer" data-auth="Verified" data-linki ndex="0">register</a> to receive the link.&nbsp\;</strong></div>\n<div><st rong> </strong>&nbsp\;</div>\n<div class="x_elementToProof"><strong>Abst ract:</strong> &nbsp\;</div>\n<div>We describe a model for a network tim e series whose evolution is governed by an underlying stochastic process\, known as the latent position process\, in which network evolution can be represented in Euclidean space by a curve\, called the Euclidean mirror. W e define the notion of a first-order changepoint for a time series of netw orks\, and construct a family of latent position process networks with und erlying first-order changepoints. We prove that a spectral estimate of the associated Euclidean mirror localizes these changepoints\, even when the graph distribution evolves continuously\, but at a rate that changes. Simu lated and real data examples on organoid networks show that this localizat ion captures empirically significant shifts in network evolution.&nbsp\;</ div>\n<div aria-hidden="true">&nbsp\;</div>\n<div><strong>Bio:</strong>  &nbsp\;</div>\n<div>Zach Lubberts is a data scientist working on the inter play of statistics and optimization\, with an emphasis on statistics on gr aphs. Some of his recent publications concern accurate estimation of the e igenvectors of random matrices and the capture of relevant signal in vario us graph models\, including time series of graphs.&nbsp\;</div>\n<div>Lubb erts earned his Ph.D. in applied mathematics and statistics from Johns Hop kins University in 2019. His dissertation research focused on the applicat ion of real algebraic geometry to the construction of multivariable tight wavelet frames for use in signal processing. He also earned his bachelor&r squo\;s degree in applied mathematics and statistics and philosophy from J ohns Hopkins in 2013.&nbsp\;</div> END:VEVENT END:VCALENDAR