Print
Univis
Search
 
FAU-Logo
Techn. Fakultät Willkommen am Institut für Informatik 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:
Friday, 12:15 - 13:45; Room: H16
Thursday, 14:15 - 15:45; Room: H16


Lecturer

Schedule of Lecture and Exercises

Summer term lecture schedules are typically somewhat frayed, due to public holidays and the Berg festival. In Pattern Analysis, we will do 2x2 lecture hours per week. This allows us to skip weeks with a public holiday (plus it keeps the Friday free after a public holiday on a Thursday). Thus, lectures take 2 hours each, exercises 2 hour each. Contents of exercise sessions of the same week are identical - it suffices to attend one of them.

Here is the full schedule:

 

WeekExercisesLectureRemarks
1-April 14, April 15
2April 18, April 19April 21, April 22
3April 28, April 29
4May 2, May 3-(public holiday on May 5)
5-May 12, May 13
6--Berg
7May 23, May 24-(public holiday on May 26)
8-June 2, June 3
9June 6, June 7June 9, June 10
10-ONLY June 16(Friday lecture cancelled)
11June 20, June 21June 23, June 24
12-June 30, July 1
13July 4, July 5July 7, July 8
14July 11, July 12July 14, July 15

 

 

 

Slides and Resources

Date

Topic
Slides / Resource
April 14, 2016

Organization

Initiates file downloadpdf
April 15, 2016Random ForestsCriminisi, Shotton: "Decision Forests for Computer Vision and Medical Image Analysis", Springer 2013, Sections 2-7.
(available online from within the university network)
Criminisi, Shotton, Konukoglu: "Decision Forests: A Unified Framework for Classification, Regression, Density Estimation, Manyfold Learning and Semi-Supervised Learning", Foundations and Trends in Computer Graphics and Vision, vol. 7, nos. 2-3, 2011, pp. 81-227.
(available online from within the university network)
April 21, 2016Probabilistic Linear FitBishop: "Pattern Recognition and Machine Learning", Springer 2006, Sec. 1.2.5-1.2.6
April 22, 2016Parzen-Rosenblatt density estimationDuda, Hart, Stork: "Pattern Classification", Wiley 2001 (2nd ed.), Section 4.3
April 29, 2016Mean ShiftComaniciu, Meer: "Mean Shift: A Robust Approach Toward Feature Space Analysis", IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 5, Mai 2002, pp. 603-619.
Download here from within the university network.
May 12, 2016ClusteringDuda, Hart, Stork: "Pattern Classification", Wiley 2001 (2nd ed.), Section 10.6-10.11
June 3, 2016Dirichlet Mixtures / Chinese Restaurant Process

Blei, Griffiths, Jordan: "The nested Chinese restaurant process and Bayesian nonparametric inference of topic hierarchies", Journal of the ACM, volume 57, no. 2, Jan. 2010, art. no. 7.
Most relevant to us are Section 1 and Section 2 of the paper.
Download Opens external link in new windowhere from within the university network.

Murphy: "Machine Learning - A Probabilistic Perspective", MIT Press 2012, Section 25.2.
Prof. Boyd-Graber put a Opens external link in new windowyoutube video online describing the core idea in a nutshell
June 16MDSA good description on multidimensional scaling by Herve Abdi can be found online Opens external link in new windowhere.
June 16, 2016IsomapTenenbaum, de Silva, C. Langford: "A Global Geometric Framework for Nonlinear Dimensionality Reduction.", Science vol. 290, no. 5500, Dec. 2000, pp. 2319-2323.
Download Opens external link in new windowhere from within the university network.
Local Linear Embedding (LLE)Roweis, Saul: "Nonlinear Dimensionality Reduction by Locally Linear Embedding", Science, vol. 290, no. 5500, Dec 2000, pp. 2323-2326.
Download Opens external link in new windowhere from within the university network.
June 23, 2016Laplacian EigenmapsBelkin, Niyogi: "Laplacian Eigenmaps for Dimensionality Reduction and Data Representation", Neural Computation, vol. 15, no. 6, June 2003, pp. 1373-1396.
Download Opens external link in new windowhere from within the university network.
Rita Osadchy gave a Opens external link in new windownice presentation on manifold learning
June 30, 2016HMMs

Bilmes: "A Gentle Tutorial of the EM Algorithm and its Application to Parameter Estimation for Gaussian Mixture and Hidden Markov Models", U.C. Berkeley, Technical Report TR-97-021, April 1997.
Download Opens external link in new windowhere.

Rabiner: "A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition", Proceedings of the IEEE, vol. 77, no. 2, Feb. 1989, pp. 257-286.
Download Opens external link in new windowhere from within the university network.
July 7, 2016MRFsGeman, Graffigne: "Markov Random Field Image Models and Their Applications to Computer Vision", International Congress of the Mathematicians, 1986.
Download Opens external link in new windowhere from within the university network.
July 14, 2016MRFs/CRFsKolmogorov, Zabih: "What Energy Functions Can Be Minimized via Graph Cuts?", IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 26, no. 2, Feb. 2004, pp. 147-159.
Download Opens external link in new windowhere from within the university network.
July 15, 2016MRFs/CRFsProf. Hamprecht put a Opens external link in new windowyoutube video online where he directly solves for the graph cut weights using a linear system of equations.