BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//TUC//Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Europe/Athens
TZNAME:EEST
DTSTART:19700329T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
TZNAME:EET
DTSTART:19701025T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CREATED:20230605T101716Z
LAST-MODIFIED:20230605T101716Z
DTSTAMP:20260418T072354Z
UID:1776486234@tuc.gr
SUMMARY:Ομιλία της Κωνσταντίνας Βαλογιάννη (
 IE Univ, Madrid) | 07.06.2023, 15:00
LOCATION:Λ - Κτίριο Επιστημών/ΗΜΜΥ, 141Π-98
DESCRIPTION:https://www.tuc.gr/el/to-polytechnei
 o/ilektronikes-ypiresies/imerologio/
 imerologio-ekdiloseon-1?tx_tucevents
 2_tuceventsdisplay%5Baction%5D=show&
 tx_tucevents2_tuceventsdisplay%5Bcon
 troller%5D=Event&tx_tucevents2_tucev
 entsdisplay%5Bevent%5D=6227&cHash=4a
 c4274d3ed1baed0fe5b30f42797510\nΤίτλ
 ος\n Causal ABM: A Methodology for L
 earning Plausible Causal Models usin
 g Agent-Based Modeling\n  \n Περίληψ
 η\n We present Causal ABM, a methodo
 logy to derive causal structures des
 cribing complex underlying behaviora
 l phenomena. Agent-based models (ABM
 s) have powerful advantages for caus
 al modeling that have not been explo
 red sufficiently. Unlike traditional
  causal estimation approaches which 
 often result in "one best" causal st
 ructure that is learned, two propert
 ies of ABMs - equifinality (the abil
 ity of different sets of conditions 
 or model representations to yield th
 e same outcome) and mutlifinality (t
 he same ABM might yield different ou
 tcomes) - can be exploited to learn 
 multiple diverse "plausible causal m
 odels" from data. Using an illustrat
 ive example of news sharing on socia
 l networks we show how this idea can
  be applied to learn such causal set
 s. We also show how genetic algorith
 ms can be used as a estimation techn
 ique to learn multiple plausible cau
 sal models from data due to their pa
 rallel search structure. However, si
 gnificant computational challenges r
 emain before this can be generally a
 pplied, and we, therefore, highlight
  specific key issues that need to be
  addressed in future work.\n \n Βιογ
 ραφικό\n Konstantina Valogianni is a
 n Assistant Professor of Information
  Systems at IE University. She has r
 eceived her PhD from Rotterdam Schoo
 l of Management, Erasmus University 
 Rotterdam (2016). Her research focus
 es on using Machine Learning to enab
 le sustainable societies. Her main l
 ine of research focuses on designing
  intelligent algorithms to facilitat
 e a better electric mobility integra
 tion in current smart grids. Her wor
 k has appeared in journals, such as 
 Information Systems Research, Produc
 tion and Operations Management, Info
 rmation & Management, Decision S
 upport Systems, Energy Policy, as we
 ll as conferences such as the Intern
 ational Conference on Information Sy
 stems (ICIS), AAAI Conference on Art
 ificial Intelligence (AAAI), the Int
 ernational Conference on Autonomous 
 Agents and Multiagent Systems (AAMAS
 ). She is teaching technology and in
 novation management and machine lear
 ning courses at the Masters and Exec
 utive levels, whereas she also teach
 es PhD courses on information system
 s.\n
STATUS:CONFIRMED
ORGANIZER;RSVP=FALSE;CN=TUC;CUTYPE=TUC:mailto:webmaster@tuc.gr
DTSTART:20230607T150000
DTEND:20230607T160000
TRANSP:OPAQUE
CLASS:DEFAULT
END:VEVENT
END:VCALENDAR