
Overview
Aims and objectives of the course
Knowledge of the most important concepts and methods form the area
adaptive filters, machine learning and neural networks.
Covered Topics
 Introduction to Machine Learning
 Learning Algorithms for Neural Networks
 Learning Algorithms and Signal Processing: adaptive
Filters
 Algorithm independent Machine Learning
 Practical Classification Algorithms
 Unsupervised Learning
 Hidden Markov Models
 Application of HMMs to Speech Recognition
Übungen / Tutorials
 Getting started
with Matlab
 Introduction to the
Matlab Neural Network Toolbox (NNT)
 Using the NNT to solve
a digit classification task
 Adaptive
filters
 Software for Classification and Regression
 Comparing classification algorithms
 Gaussian
Statistics
 PCA, ICA, Blind Source
Separation
 Hidden Markov
Models
 Mixtures of
Gaussians
 Automatic Speech
Recognition
Prerequisites
There are no particular courses which must be taken as
prerequisites for this course. Although there will be an
introduction to MATLAB in the beginning of the exercises, it is
recommended to have already some basic knowledge and experience in
it. We also assume elementary mathematical knowledge in probability
theory, statistics, analysis and calculus.
Bibliography and Teaching aids
 [DHS] R. O. Duda, P.E. Hart, D. G. Stork: Pattern
Classification, 2nd edition, John Wiley & Sons, 2001. (The publisher provides all figures in PDF format as well as a list of errata) 30 copies are available
in the
LBS .
 [BIS06] C. Bishop: Pattern Recognition and Machine Learning, Springer,
2006, 12 copies are available in the
LBS .
 [BIS] C. Bishop: Neural Networks for Pattern Recognition, Oxford Univ.
Press, 1996.
 [HAY] S. Haykin: Neural Networks: A Comprehensive Foundation,
Macmillan/IEEE Press, 1994, 8 copies available in the
LBS
 [HKP] J. Hertz, A. Krogh, and R. G. Palmer. Introduction to the Theory of Neural Computation.
Addison Wesley, 1991. [amazone]
 [NNT] Documentation (pdf) to the Matlab Neural Network Toolbox.
 [HA1] S. Haykin: Adaptive Filter Theory, Third
Edition, PrenticeHall, Inc., Upper Saddle River, NJ, 1996.
 [WS] B. Widrow and S. D. Stearns: Adaptive Signal
Processing, PrenticeHall, Inc., Upper Saddle River, NJ, 1985.
 [BNT] K. Murphy: Bayes Net Toolbox for Matlab.
 [BIL] J. Bilmes: A Gentle Tutorial on the EM Algorithm and its Application to
Parameter Estimation for Gaussian Mixture and Hidden Markov
Models, Technical Report, University of Berkeley,
ICSITR97021, 1997.
 [VB] VOICEBOX: Speech Processing Toolbox for
MATLAB.
 [RAB] L. R. Rabiner: A tutorial on hidden Markov models and selected
applications in speech recognition, Proceedings of the IEEE, 77
(2), 257286.
 [HTK] S. Young, G. Evermann, T. Hain, D. Kershaw, G.
Moore, J. Odell, D. Ollason, D. Povey, V. Valtchev, P. Woodland:
The HTK
Book.
 [HAH] X. Huang, A. Acero, H.Hon: Spoken Language
processing: A Guide to Theory, Algorithm and System Development,
Prentice Hall, Redmond, Washington, 2001.
 [RJ] L. R. Rabiner, B. Juang: Fundamentals of Speech
Recognition, PTR Prentice Hall (Signal Processing Series),
Englewood Cliffs NJ, 1993, ISBN 0130151572.
 [GM] B.Gold, N. Morgan: Speech and audio signal
processing: processing and perception of speech, and music, John
Wiley and Sons Inc., 1999, ISBN: 0471351547.

