Print
Univis
Search
 
FAU-Logo
Techn. Fakultät Website deprecated and outdated. Click here for the new site. FAU-Logo

Pattern Analysis [PA]

Summary
This lecture complements (and builds on top of) the lectures "Introduction to Pattern Recognition" and "Pattern Recognition". In this third edition, we focus on modeling of densities, and how to use these models for analyzing the data. Major topics of this lecture are regression, density estimation, manifold learning, hidden Markov models, conditional random fields, and random forests. The lecture is accompanied by exercises, where theoretical results are practically implemented and applied.

Dates & Rooms:
Wednesday, 10:15 - 11:45; Room: H16
Monday, 16:15 - 17:45; Room: H16


Lecturer

Organization

Pattern Analysis gives 5 ECTS points, which consists of the equivalent of 3 hours of lecture, and 1 hour of exercises.

You need to pass both the exercises and the lecture to obtain the 5 ECTS certificate. To pass the exercises, you work in a group of two (i.e., you and your partner), and you need to pass each of the five assignments by showing and discussing your results with one of the supervisors of the exercises. If you passed the exercises, you can do the oral exam about lecture+exercises. The outcome of this oral exam is the grade for the class.

Please enroll to the class in studon, for two reasons: a) to obtain the work sheets and lecture materials, b) to give us an opportunity to contact you via email if something unforeseen happens, e.g., a class has to be canceled due to sickness.

Schedule

Lecture: We reserved two time slots per week for the lecture, with two hours each. This leads to a somewhat compressed schedule that gives you a little bit of breathing space towards the end of the semester - perfect if you'd like to do the exam in the first examination period.

Exercises: There are four exercise slots per week, namely Tuesday 2pm-3pm, 3pm-4pm in 02.151-113 (2nd floor of the blue CS tower), and Thursday 4pm-5pm, 5pm-6pm in 01.142 (Huber Cip), 00.0156-113, and 02.151b-113 (ground level and 2nd floor CS tower). Please choose one for you and your exercise partner.

 

Schedule:

 

Monday LectureWednesday LectureTuesday ExerciseThursday Exercise
[no lecture: Easter monday]April 24--
April 29[no lecture: May 1st]April 30: warmupMay 2: warmup
[CANCELLED]May 8May 7May 9
May 13May 15May 14May 16
May 20May 22May 21May 23
May 27May 29--
June 3June 5June 4June 6
[no lecture: Pentecost monday][no lecture: Berg hangover]-June 13
June 17June 19June 18-
June 24June 26June 25June 27
July 1July 3 July 2July 4
--July 9July 11
July 15July 17: wrapup, repetitionJuly 16July 18
--July 23July 25

 

Remarks on the schedule:

  • The lecture ends already on July 15, but there is one additional meeting two weeks later on July 17 for a small repetition, some hints on the oral exam, and room for your questions.
  • The exercise "warmup" in the second week is perfect for you if you have never done image processing in python, in order to get familiar with some basic tools and commands.
  • We try to keep exercises synchronized between Tuesday and Thursday, although there are two public holidays on Thursdays, and the Berg week. Typically, the same material is presented within the same week, with the notable exception that Thursday June 13 and Tuesday June 18 are also one pair.

 

 

Exams

You can do the exam in the first examination period (July 29 - August 10) or in the second examination period (September 20 - October 12). The offered examination dates are:

 

Possible Date(s)Remarks
Mo. July 291st examination period
Fr. Aug. 2 - Fr. Aug. 91st examination period
Mo. Aug. 19 - Fr. Aug. 23(bookable, but outside of official examination periods)
Thu. Sep. 19(bookable, but outside of official examination periods)
Fr. Sep. 20 - Fr. Oct. 112nd examination period

 

I the lecture, you can sign up for you preferred examination date from June 24 on. The distribution of available dates is done on a first-come-first-served basis. If, for some reason, it is not possible for you to register for an examination date in the lecture, please send me an email with your three preferred dates at June 27 or later.

Once you registered for a date, it is considered fixed (unless there are urgent reasons outside of your control). I am expecting a large volume of oral exams this year - so please consider to do the exam prior to the second examination period. If you do not make an appointment with me until Sep. 5, an examination date will be assigned to you in meinCampus.