SluitenHelpPrint
Switch to English
Cursus: INFOPROB
INFOPROB
Probabilistic reasoning
Cursus informatieRooster
CursuscodeINFOPROB
Studiepunten (ECTS)7,5
Categorie / NiveauM (Master)
CursustypeCursorisch onderwijs
VoertaalEngels
Aangeboden doorFaculteit Betawetenschappen; Graduate School of Natural Sciences;
Contactpersoondr. S. Renooij
Telefoon+31 30 2539266
E-mailS.Renooij@uu.nl
Docenten
Docent
dr. S. Renooij
Feedback en bereikbaarheid
Overige cursussen docent
Blok
1-GSNS  (04-09-2017 t/m 10-11-2017)
Aanvangsblok
1-GSNS
TimeslotD: WO-middag, WO-namiddag, Vrijdag
Onderwijsvorm
Voltijd
Cursusinschrijving geopendvanaf 29-05-2017 t/m 25-06-2017
AanmeldingsprocedureOsiris
Inschrijven via OSIRISJa
Inschrijven voor bijvakkersJa
VoorinschrijvingNee
Na-inschrijvingJa
Na-inschrijving geopendvanaf 21-08-2017 t/m 17-09-2017
WachtlijstJa
Plaatsingsprocedureadministratie onderwijsinstituut
Cursusdoelen
After completing the course, the student
  • recognises the strengths and weaknesses of the probabilistic graphical model framework for reasoning under uncertainty
  • understands the relation between probabilistic independence and the graphical representation of a probabilistic network
  • can compute probabilities from a probabilistic network
  • understands Pearl’s algorithm for probabilistic inference
  • understand the different patterns of reasoning allowed in probabilistic networks
  • knows about the challenges of constructing probabilistic networks for real applications
  • understands the benefits and drawbacks of parameterized models, and the details of Noisy-Or
  • understands a basic algorithm for automated network construction from data
  • knows the purpose of sensitivity analysis and understands the related functions and concepts
Inhoud
Human experts have to make judgments and decisions based on uncertain, and often even conflicting, information. To support these complex decisions, knowledge-based systems should be able to cope with this type of information. Probability theory is one of the oldest theories dealing with the concept of uncertainty. In this course, probabilistic models for manipulating uncertain information in knowledge-based systems are considered. More specifically, the theory underlying the framework of probabilistic networks are considered, and the construction of such networks for real-life applications are discussed.
Ingangseisen
Je moet een geldige toelatingsbeschikking hebben
Verplicht materiaal
-
Aanbevolen materiaal
Syllabus
Syllabus 'Probabilistic Reasoning', available at the 'Studiepunt' in MIN 911;
Handleiding
Studymanual, available online (see website with additional information);
Artikelen
Course transparencies, also available online.
Software
Geen software nodig
Werkvormen (aanwezigheidsplicht)
Hoorcollege

Toetsen
Eindresultaat
Weging100
Minimum cijfer-

SluitenHelpPrint
Switch to English