392118 Vertiefung Maschinelles Lernen (Ü) (WiSe 2014/2015)
Contents, comment
-
Aufbauend auf dem Grundlagen-Modul "Neuronale Netze und Lernen", welches die grundlegende Theorie des maschinellen Lernens sowie einige grundlegende Ansätze behandelt hat, werden in diesem Modul weitere, komplexere Lernarchitekturen behandelt. Die Themen der Vorlesung umfassen insbesondere:
Ensemble-Verfahren - gewichtete Kombination mehrerer Lern-Module
Mixture-of-Experts - (hierarchische) Zuweisung von Subproblemen zu Experten-Modulen
Aktives Lernen
Reinforcement-Lernen
Partially Observable Markov Decision Processes (POMDPs)
Gaussian Processes: Bishop, Kapitel 6.4
Graphical Models
Sampling: Bishop, Kapitel 11Übungen
Anstatt wöchentliche Übungszettel zu bearbeiten, sollen Sie bis Ende November eine Projektaufgabe realisieren: Suchen Sie eine Implementierung des Viola-Jones-Algorithmus heraus und wenden Sie ihn auf ein komplexes Klassifikationsproblem an. In Frage kommen z.B. Gesichtserkennung, Handerkennung oder sogar Hand-Postur-Erkennung, d.h. Klassifikation der Handstellung.
Datenbanken von Gesichtsbildern
Datenbank von Handposturen
Handposturerkennung mit AdaBoost und SIFT-Features (Full Text)http://ni.www.techfak.uni-bielefeld.de/teaching/vertiefung-maschinelles-lernen
Requirements for participation, required level
-
Die Vorlesung wendet sich an einschlägig interessierte Studenten der Informatik, Mathematik und Linguistik im Hauptstudium. Neuronale Netze und Lernen
Bibliography
-
Bishop, Ch., "Pattern Recognition and Machine Learning", Springer
Mitchel, T., "Machine Learning",
Viola, P., Jones, M., "Robust Real-Time Face Detection", International Journal of Computer Vision 57(2), 137–154, 2004
Sutton & Barto, "Reinforcement Learning: An Introduction", MIT Press
Vorlesungsfolien POMDPs, W. Burgard, Uni Freiburg
David MacKay: "Gaussian Processes Basics" (video lecture)
Iain Murray: "Markov Chain Monte Carlo" (video lecture)
External comments page
Teaching staff
Dates ( Calendar view )
| Frequency | Weekday | Time | Format / Place | Period |
|---|
Subject assignments
| Module | Course | Requirements | |
|---|---|---|---|
| 39-M-Inf-VML Vertiefung Maschinelles Lernen | Vertiefung Maschinelles Lernen | Graded examination
|
Student information |
The binding module descriptions contain further information, including specifications on the "types of assignments" students need to complete. In cases where a module description mentions more than one kind of assignment, the respective member of the teaching staff will decide which task(s) they assign the students.
| Degree programme/academic programme | Validity | Variant | Subdivision | Status | Semester | LP | |
|---|---|---|---|---|---|---|---|
| Intelligente Systeme / Master | (Enrollment until SoSe 2012) | Vertiefung Maschinelles Lernen | Wahlpflicht | 1. | 5 | benotet | |
| Naturwissenschaftliche Informatik / Diplom | (Enrollment until SoSe 2004) | Robotik; Physik; Biologie; NNet; ME | HS | ||||
| Naturwissenschaftliche Informatik / Master | (Enrollment until SoSe 2012) | Vertiefung Maschinelles L | Wahlpflicht | 1. | 5 | benotet |
- Registered number: 10
- This is the number of students having stored the course in their timetable. In brackets, you see the number of users registered via guest accounts.
- Address:
- WS2014_392118@ekvv.uni-bielefeld.de
- This address can be used by teaching staff, their secretary's offices as well as the individuals in charge of course data maintenance to send emails to the course participants. IMPORTANT: All sent emails must be activated. Wait for the activation email and follow the instructions given there.
- If the reference number is used for several courses in the course of the semester, use the following alternative address to reach the participants of exactly this: VST_49054704@ekvv.uni-bielefeld.de
- Coverage:
- 2 Students to be reached directly via email
- Notes:
- Additional notes on the electronic mailing lists
- Last update basic details/teaching staff:
- Friday, December 11, 2015
- Last update times:
- Thursday, September 18, 2014
- Last update rooms:
- Thursday, September 18, 2014
- Type(s) / SWS (hours per week per semester)
- exercise (Ü) / 1
- Department
- Faculty of Technology
- Questions or corrections?
- Questions or correction requests for this course?
- Planning support
- Clashing dates for this course
- Links to this course
- If you want to set links to this course page, please use one of the following links. Do not use the link shown in your browser!
- The following link includes the course ID and is always unique:
- https://ekvv.uni-bielefeld.de/kvv_publ/publ/vd?id=49054704
- Send page to mobile
-
Click to open QR code
Scan QR code:
- ID
- 49054704