BEGIN:VCALENDAR VERSION:2.0 PRODID:ILLC Website X-WR-TIMEZONE:Europe/Amsterdam BEGIN:VTIMEZONE TZID:Europe/Amsterdam X-LIC-LOCATION:Europe/Amsterdam BEGIN:DAYLIGHT TZOFFSETFROM:+0100 TZOFFSETTO:+0200 TZNAME:CEST DTSTART:19700329T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:+0200 TZOFFSETTO:+0100 TZNAME:CET DTSTART:19701025T030000 RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU END:STANDARD END:VTIMEZONE BEGIN:VEVENT UID:/NewsandEvents/Archives/2003/newsitem/530/22-O ctober-2003-Learning-Solutions-2003-Adaptive-Intel ligence-in-research-and-practical-applications-Rad boud-Auditorium-Nijmegen-the-Netherlands DTSTAMP:20030926T000000 SUMMARY:Learning Solutions 2003: Adaptive Intellig ence in research and practical applications, Radbo ud Auditorium, Nijmegen, the Netherlands DTSTART;VALUE=DATE:20031022 DTEND;VALUE=DATE:20031022 LOCATION:Radboud Auditorium, Nijmegen, the Netherl ands DESCRIPTION:Neural networks are computer programs that are able to learn. Their functioning is inspi red by the function of the brain. The value added by neural networks is strongest for those problems that lack explicit knowledge. A large number of n eural network aided applications has already been realized. Well-known applications are pattern reco gnition, time series prediction, and process contr ol. Neural networks do not always produce the best solution, however. Better solutions are therefore often obtained through a combination with explici t domain knowledge. Bayesian statistics offers an elegant formalism to combine learning and explicit modeling. Furthermore, statistical methods for qu antification of reliability are of great importanc e. A modern trend is therefore marked by an integr ated approach that combines neural networks with m ethods from statistics and artificial intelligence . The symposium 'Learning Solutions 2003' offers a n up-to-date overview of Dutch research in this ar ea. For more information, see http://www.snn.kun .nl/nederland/index.php3?page=15 X-ALT-DESC;FMTTYPE=text/html:\n
\nNeural n etworks are computer programs that are able to lea rn.\nTheir functioning is inspired by the function of the brain. \nThe value added by neural network s is strongest for those \nproblems that lack expl icit knowledge.\n A large number of neural network aided applications has already been realized. \nW ell-known applications are pattern recognition, ti me series prediction,\n and process control.\n Neu ral networks do not always produce the best soluti on, however. \nBetter solutions are therefore ofte n obtained through a combination\n with explicit d omain knowledge.\n Bayesian statistics offers an e legant formalism to combine learning \nand explici t modeling. \nFurthermore, statistical methods for quantification of reliability \nare of great impo rtance. \nA modern trend is therefore marked by an integrated approach that \ncombines neural networ ks with methods from statistics and \nartificial i ntelligence. \nThe symposium 'Learning Solutions 2 003' offers an up-to-date \noverview of Dutch rese arch in this area.\n
\n \nFor more information, see\n http://www.snn.kun.nl/nederland/index.php 3?page=15\n
URL:/NewsandEvents/Archives/2003/newsitem/530/22-O ctober-2003-Learning-Solutions-2003-Adaptive-Intel ligence-in-research-and-practical-applications-Rad boud-Auditorium-Nijmegen-the-Netherlands END:VEVENT END:VCALENDAR