Inhalt

[ 875BIN2MUTU13 ] UE Machine Learning: Unsupervised Techniques

Versionsauswahl
Es ist eine neuere Version 2016W dieser LV im Curriculum Masterstudium Computer Science 2016W vorhanden.
Workload Ausbildungslevel Studienfachbereich VerantwortlicheR Semesterstunden Anbietende Uni
1,5 ECTS M1 - Master 1. Jahr Informatik Sepp Hochreiter 1 SSt Johannes Kepler Universität Linz
Detailinformationen
Quellcurriculum Masterstudium Bioinformatics 2013W
Ziele This practical course complements the lecture "Machine Learning: Unsupervised Techniques" and aims at practicing the concepts and methods acquired in the lecture.
Lehrinhalte
  • Error models
  • Information bottleneck
  • Maximum likelihood and the expectation maximization algorithm
  • Maximum entropy methods
  • Basic clustering methods, hierarchical clustering, and affinity propagation
  • Mixture models
  • Principal component analysis, independent component analysis, and other projection methods
  • Factor analysis
  • Matrix factorization
  • Auto-associator networks and attractor networks
  • Boltzmann and Helmholtz machines
  • Hidden Markov models
  • Belief networks
  • Factor graphs
Beurteilungskriterien Marking is based on homework
Lehrmethoden Students are given assignments in 1-2 week intervals. Homework must be handed in. Results are to be presented and discussed in the course.
Abhaltungssprache Englisch
Literatur Assignments and homework submissions are managed via JKU Moodle. Where necessary, complimentary course material is provided for download.
Lehrinhalte wechselnd? Nein
Äquivalenzen in collaboration with 675MLDAMSTU13: UE Machine Learning: Supervised Techniques (1,5 ECTS) equivalent to
875BIN2TMLU12: UE Theoretical Bioinformatics and Machine Learning (3 ECTS) -or- BIMPHUEBIN2: UE Bioinformatik II: Theoretische Bioinformatik und Maschinelles Lernen (3 ECTS)
Präsenzlehrveranstaltung
Teilungsziffer 35
Zuteilungsverfahren Direktzuteilung