BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Guatemala
BEGIN:STANDARD
DTSTART:20060930T230000
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20241020T193427Z
UID:F635D157-65F8-442D-8542-DD2E1CEA14DE
DTSTART;TZID=America/Guatemala:20241017T150000
DTEND;TZID=America/Guatemala:20241017T170000
DESCRIPTION:Dr. Justin Dauwels is an Associate Professor at TU Delft (Signa
 ls and Systems\, Department of Microelectronics) and serves as co-Director
  of the Safety and Security Institute at TU Delft. He is also the scientif
 ic lead of the Model-Driven Decisions Lab (MoDDL)\, the first lab for the 
 Knowledge Building program between the Netherlands police and TU Delft.\n\
 nHis research interests include data analytics with applications to intell
 igent transportation systems\, autonomous systems\, and the analysis of hu
 man behavior and physiology. His academic lab has spawned four startups ac
 ross various industries\, ranging from AI for healthcare to autonomous veh
 icles.\n\nTitle: Object-centric Generative AI\n\nAbstract:\nGenerative AI 
 refers to a category of artificial intelligence models designed to generat
 e new content\, such as text\, images\, audio\, or other types of data. Pr
 obably the best-known example of generative AI is ChatGPT\, the fastest co
 nsumer application to reach 100 million monthly active users. Generative A
 I models use machine learning algorithms to learn patterns and structures 
 from existing data\, then produce new data that is similar in style or con
 tent to the training data.\n\nIn this presentation\, I will begin with a t
 utorial on various approaches to generative AI. I will then discuss projec
 ts in our group at TU Delft focused on deep generative models\, specifical
 ly object-centric generative models. I will briefly introduce novel deep g
 enerative models developed by our team. Next\, I will explain how we are a
 pplying these models for rainfall nowcasting\, integrating physical laws i
 nto the deep generative models. Finally\, I will highlight various AI-rela
 ted initiatives our team is involved in.\n\nSpeaker(s): Dr. Justin Dauwels
 \, \n\nBldg: Miniauditorio de la Facultad de Ingeniería\, UCR Escuela de 
 Ingeniería Eléctrica\, San Pedro\, San Jose\, Costa Rica
LOCATION:Bldg: Miniauditorio de la Facultad de Ingeniería\, UCR Escuela de
  Ingeniería Eléctrica\, San Pedro\, San Jose\, Costa Rica
ORGANIZER:costarica-sps@ieee.org
SEQUENCE:2
SUMMARY:Object-centric Generative AI
URL;VALUE=URI:https://events.vtools.ieee.org/m/440156
X-ALT-DESC:Description: <br /><p><strong>Dr. Justin Dauwels</strong> is an 
 Associate Professor at TU Delft (Signals and Systems\, Department of Micro
 electronics) and serves as co-Director of the Safety and Security Institut
 e at TU Delft. He is also the scientific lead of the Model-Driven Decision
 s Lab (MoDDL)\, the first lab for the Knowledge Building program between t
 he Netherlands police and TU Delft.</p>\n<p>His research interests include
  data analytics with applications to intelligent transportation systems\, 
 autonomous systems\, and the analysis of human behavior and physiology. Hi
 s academic lab has spawned four startups across various industries\, rangi
 ng from AI for healthcare to autonomous vehicles.</p>\n<h3><strong>Title</
 strong>: Object-centric Generative AI</h3>\n<p><strong>Abstract</strong>:<
 br>Generative AI refers to a category of artificial intelligence models de
 signed to generate new content\, such as text\, images\, audio\, or other 
 types of data. Probably the best-known example of generative AI is ChatGPT
 \, the fastest consumer application to reach 100 million monthly active us
 ers. Generative AI models use machine learning algorithms to learn pattern
 s and structures from existing data\, then produce new data that is simila
 r in style or content to the training data.</p>\n<p>In this presentation\,
  I will begin with a tutorial on various approaches to generative AI. I wi
 ll then discuss projects in our group at TU Delft focused on deep generati
 ve models\, specifically object-centric generative models. I will briefly 
 introduce novel deep generative models developed by our team. Next\, I wil
 l explain how we are applying these models for rainfall nowcasting\, integ
 rating physical laws into the deep generative models. Finally\, I will hig
 hlight various AI-related initiatives our team is involved in.</p>
END:VEVENT
END:VCALENDAR