BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:Eventi DIAG
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20191027T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20190331T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RDATE:20200329T020000
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.18751.field_data.0@www.open.diag.uniroma1.it
DTSTAMP:20260410T210441Z
CREATED:20190919T174558Z
DESCRIPTION:Modelling human-based phenomena is not an easy task. Moreover\,
  implementing such a model by means of a traditional algorithmic approach 
 - based on ab initio assumptions - gets nearly impossible.On the other han
 d\, leaving the idea of any prodromal hypothesis\, it is possible to let m
 odels emerge by themselves. That can be done by applying specific mathemat
 ical methods and several advanced tools such as feature representation tra
 nsforms and neural networks with a little pragmatic perspective on chaotic
  phenomena. Sometimes results gets as surprising as unexpected\, but\, eve
 ntually\, quite interesting. The interpretation of apparently chaotic phen
 omena relies on our ability to extrapolate symmetries and mathematical rel
 ations\, therefore it depends on how we represent the information that\, w
 hile apparently hidden in the initial form\, can be also described in a di
 fferent manner. When we change frame - or basis\, in a strictly mathematic
 al fashion - we can decompose a signal in new sets of coefficients that co
 uld dramatically change our perspective. Signal decomposition techniques s
 uch as functional analysis\, multiresolution transforms and polynomial dec
 ompositions\, permits us to pack the information contained on a signal int
 o a few significant numerical  coefficients. By doing so we could relate f
 ew coefficients to specific subsets of features emerging on the observed p
 henomena although initially unfathomable on the registered signal. This ki
 nd of informational assets is generally ideal for automatic classification
  and model extrapolation by means of machine learning techniques and neura
 l networks.Such models can be applied to many fields of research to unders
 tand and exploit non periodical phenomena as well as for classification pu
 rposes in many different fields. Such approaches constitutes optimal tools
  in order to understand\, model and predict the behaviour of human groups 
 as well as to better understand and predict important aspects of human-com
 puter interaction.In this talk the following cues will be mentioned:      
   - Predictive and forecasting neural models        - Dynamical time-evolv
 ing AI-based models        - Continuous machine learning approaches       
  - AI-oriented feature enhancement and suppression        - Non-trivial hu
 man-based classification tasks        - Context and environment driven neu
 ro-classifiers        - Behavioural models identification and predictive i
 nteraction        - A new biometrics perspective and other things to come.
 ... 
DTSTART;TZID=Europe/Paris:20190924T091500
DTEND;TZID=Europe/Paris:20190924T091500
LAST-MODIFIED:20191008T082735Z
LOCATION:Aula A6
SUMMARY:AI: Advances in Imagination - Christian Napoli
URL;TYPE=URI:http://www.open.diag.uniroma1.it/node/18751
END:VEVENT
END:VCALENDAR