University of Leiden, LIACS (October-November, 2007)

Connectionist Density Estimation for Structures and Sequences

Edmondo Trentin

 


Timetable: Friday 11.15-13.00, room 402.

Slides of the course (in .pdf format). Note: this file will be updated from time to time with new stuff.

Textbooks and Papers:

1. Duda & Hart, "Pattern Classification and Scene Analysis". J. Wiley, 1973 (equivalentemente: Duda, Hart & Stork, "Pattern Classification - Second Edition". J. Wiley, 2001).

2. L. R. Rabiner, "A tutorial on hidden Markov models and selected applications in speech recognition". Proc. of IEEE, vol. 77, no. 2, pp. 257--286, February 1989.

3. E. Trentin & M. Gori, "A survey of hybrid ANN/HMM models for automatic speech recognition". Neurocomputing, 37(1/4); 91-126, March 2001.

4. E. Trentin & M. Gori, "Robust Combination of Neural Networks and Hidden Markov Models for Speech Recognition". IEEE Transactions on Neural Networks, vol. 14, n. 6; 1519-1531, Nov. 2003.

5. C. Bishop, "Neural Networks for Pattern Recognition". Oxford University Press, 1995.

6. C. Bishop, "Pattern Recognition and Machine Learning". Springer, 2006.

7. E. Trentin, "Simple and Effective Connectionist Nonparametric Estimation of Probability Density Functions". Proceedings of ANNPR 2006, 1-10, Ulm, Germany, Aug 31-Sep 2, 2006.

8. E. Trentin and E. Di Iorio, "Unbiased SVM Density Estimation with Application to Graphical Pattern Recognition". Proceedings of ICANN 2007, (II)271-280, Porto (Portugal), Sep. 2007.


UCI Machine Learning Repository of benchmark datasets with pattern classification problems.

Web page with a list of neural network simulation software for different operating systems.


Course outline

1. The problem of probability estimation for the "optimal" classification of real-valued patterns

2. Review of neural nets; Multilayer Perceptron (MLP) and the backprop algorithm

3. MLPs for the supervised estimation of class posterior probabilities

4. Unsupervised nonparametric {\em pdf} estimation via Parzen Neural Networks (PNN)

5. An effective Bayesian approach to the classification of structured data via PNN

6. Review of hidden Markov models (HMM)

7. Hybrid MLP/HMM systems for sequences


Edmondo Trentin


Back to the LIACS home page.

(Last updated: Oct 09, 2007)