BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Asia/Bangkok
BEGIN:DAYLIGHT
DTSTART:20380119T101407
TZOFFSETFROM:+0700
TZOFFSETTO:+0700
RRULE:FREQ=YEARLY;BYDAY=3TU;BYMONTH=1
TZNAME:+07
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:19200401T001756
TZOFFSETFROM:+0642
TZOFFSETTO:+0700
RRULE:FREQ=YEARLY;BYDAY=1TH;BYMONTH=4
TZNAME:+07
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20200101T020718Z
UID:82167F75-5230-4528-B744-6835A2F7AE70
DTSTART;TZID=Asia/Bangkok:20191015T110000
DTEND;TZID=Asia/Bangkok:20191015T120000
DESCRIPTION:Data Fusion Through Matrix and Tensor Decompositions:\nAn Overv
 iew of Solutions\, Challenges\, and Prospects\n\nIn many fields today\, su
 ch as neuroscience\, remote sensing\, computational social science\, and p
 hysical sciences\, multiple sets of data are readily available. The datase
 ts might either be multimodal where information about a given phenomenon i
 s obtained through different types of acquisition techniques resulting in 
 datasets with complementary information but essentially of different types
 \, or multiset where the datasets are all of the same type but acquired fr
 om different samples\, at different time points\, or under different condi
 tions.\n\nMatrix and tensor factorizations enable joint analysis\, i.e.\, 
 fusion\, of these multiple datasets such that they can fully\ninteract and
  inform each other while also minimizing the assumptions placed on their i
 nherent relationships.\nA key advantage of these methods is the direct int
 erpretability of their results. This talk presents an overview\nof the mai
 n models that have been successfully used for fusion of multiple datasets\
 , in particular those\nthat are based on independent component and vector 
 analysis as well as canonical polyadic decomposition.\nExamples are presen
 ted to demonstrate their effectiveness\, and the main challenges and the o
 pportunities\nin the area are also addressed.\n\nCo-sponsored by: Chulalon
 gkorn University\n\nSpeaker(s): Tulay Adali\, \n\nDepartment of Electrical
  EngineeringChulalongkorn Unilversity254 Phayathai RoadBangkok 10330 Thail
 and\, Bangkok\, Bangkok Metropolis\, Thailand\, 10330
LOCATION:Department of Electrical EngineeringChulalongkorn Unilversity254 P
 hayathai RoadBangkok 10330 Thailand\, Bangkok\, Bangkok Metropolis\, Thail
 and\, 10330
ORGANIZER:supavadee.a@chula.ac.th
SEQUENCE:1
SUMMARY:IEEE Signal Processing Society Distinguished Lecture
URL;VALUE=URI:https://events.vtools.ieee.org/m/216893
X-ALT-DESC:Description: <br /><p>Data Fusion Through Matrix and Tensor Deco
 mpositions:&nbsp\;<br />An Overview of Solutions\, Challenges\, and Prospe
 cts</p>\n<p>&nbsp\;</p>\n<div>In many fields today\, such as&nbsp\;&nbsp\;
 neuroscience\, remote sensing\, computational social science\, and physica
 l sciences\,&nbsp\;multiple sets of data are readily available. The datase
 ts might either be&nbsp\;multimodal&nbsp\;where information about a given&
 nbsp\;phenomenon is obtained through different types of acquisition techni
 ques resulting in datasets with complementary&nbsp\;information but essent
 ially of different types\, or&nbsp\;multiset&nbsp\;where the datasets are 
 all of the same type but acquired from&nbsp\;different samples\, at differ
 ent time points\, or under different conditions.&nbsp\;<br /><br />Matrix 
 and tensor factorizations enable joint analysis\, i.e.\,&nbsp\;fusion\,&nb
 sp\;&nbsp\;of these multiple&nbsp\;&nbsp\;datasets such that they can full
 y&nbsp\;<br />interact and inform each other while also minimizing the ass
 umptions placed on their inherent relationships.&nbsp\;</div>\n<div>A key 
 advantage of these methods is the direct interpretability of their results
 .&nbsp\;This talk presents an overview&nbsp\;</div>\n<div>of the main mode
 ls that have been successfully used for fusion of multiple datasets\,&nbsp
 \;in particular those&nbsp\;</div>\n<div>that are based on independent com
 ponent and vector analysis as well as canonical polyadic&nbsp\;decompositi
 on.<br />Examples are presented to demonstrate their effectiveness\, and t
 he&nbsp\;main challenges and the opportunities&nbsp\;</div>\n<div>in the a
 rea are&nbsp\;also addressed.</div>
END:VEVENT
END:VCALENDAR