Seminar Computational Intelligence F (708.116)

WS 2005/06

Institut für Grundlagen der Informationsverarbeitung (708)

Lecturer: O.Univ.-Prof. Dr. Wolfgang Maass

Office hours: by appointment (via e-mail)


Location: IGI-seminar room, Inffeldgasse 16b/I, 8010 Graz
Date: starting from Oct. 10, 2005 every Monday, 16.15 - 18.00 p.m.

Content of the seminar:

This Seminar will examine the current state of the art in nonlinear time series analysis, with an emphasis on methods from machine learning (and an emphasis on classification, indexing, clustering, dimension-reduction, and pattern extraction;  rather than times series prediction).

In particular, we will discuss of methods for defining a suitable metric on time series,  kernel based methods, neural net based methods, methods of dynamic bayesian networks, and methods for benchmarking such algorithms. This seminar will be carried out jointly with Horst Bischof and Gernot Kubin, since one goal of this seminar is to examine learning methods that appear to be relevant for ongoing research in the MISTRAL project. This is a joint research project of many computer science institutes at our university. (MISTRAL stands for Measurable intelligent and secure semantic extraction and retrieval of multimedia data).


Talks will be assigned at our first meeting on Oct. 10.
Talks will probably have a length of  about 40 minutes.

Schedule of talks:


Malte Rasch (?) - "Probabilistic Discovery of Time Series Motifs"
B Chiu, E Keogh, S Lonardi - the 9th ACM SIGKDD International Conference on Knowledge, 2003

Gerhard Neumann - "Gaussian Process Dynamical Models"
Jack M. Wang, David J. Fleet, Aaron Hertzmann
Presentation:: (PPT)

Andreas Juffinger - "On the Need for Time Series Data Mining Benchmarks: A Survey and Empirical Demonstration"
E Keogh, S Kasetty - Data Mining and Knowledge Discovery, 2003

Stefan Häusler - "Making Time-series Classification More Accurate Using Learned Constraints"
CA Ratanamahatana, E Keogh - proceedings of SIAM International Conference on Data Mining, 2004

Tuan van Pham - "HMM-models + HTK -models + applications in speech processing"
Lawrence R. Rabiner,  "A Tutorial in Hidden Markov Models and Selected Applications in Speech Recognition",
Proc. of the IEEE, 77(2):257-286, 1989
P. Smyth, D. Heckerman, M.I. Jordan, "Probabilistic Independence Networks for Hidden Markov Probability Models",
Technical Report, Microsoft Research, June, 1996
P. Smyth, "Belief networks, hidden Markov models, and Markov random fields: a unifying view", Pattern Recognition Letters, 1998

K. P. Murphy, "Dynamic Bayesian Networks: Representation, Learning and Inference", PhD. thesis, University of California, Berkeley, 2002
The Hidden Markov Model Toolkit (HTK),, Oct.~2005.

Marian Kepesi - "Bayesian filtering + applications of Bayesian methods in speech recognition for location estimation"
D. Fox, J. Hightower, L. Liao, D. Schulz, and G. Borriello,
"Bayesian filtering for location estimation," IEEE Pervasive Computing, 2(3):24-33, July-September 2003.

Ashley Mills - "Learning Dynamic Bayesian Networks"
Z Ghahramani - Summer School on Neural Networks, 1997
Presentation: (PPT)

David Sussillo on material from
"Kernel methods for time series classification"
Using support vector machines for time series prediction,
Klaus-Robert Müller, Alexander J. Smola , G Raetsch, B Schoekopf, J Kohlmorgen, V Vapnik
Advances in kernel methods: support vector learning table, 1999

SVM Kernels for Time Series Analysis,  S Ruping -

Dynamical Modeling with Kernels for Nonlinear Time Series Prediction L Ralaivola, FA Buc - 2003


Martin Bachler on material from
Kevin Murphy, "Dynamic Bayesian Networks: Representation, Inference and Learning"
PhD Thesis, UC Berkeley, Computer Science Division, July 2002.

Learning the Structure of Dynamic Probabilistic Networks
Nir Friedman, Kevin Murphy, and Stuart Russell.
/UAI '98 (Uncertainty in AI)./

Schwarz, Estimating the dimension of a model. Annals of Statistics, 6, 1978.

Tuesday (!!!), 31.01.2006, 16:15

Presentation of current versions of benchmark datasets in MISTRAL,
discussion of the learning tasks involved, and possible methods to be used.

                                                                                                                                                                                                                                                                                                      2006-01-20, daniela