Παρουσίαση/Προβολή

(TE4) -  Νικολαος Μητιανουδης

Περιγραφή Μαθήματος

Signal Processing and Machine Learning (TE4)

 

This course covers the basic elements of Signal Processing and Machine Learning. The course starts from the analog signal domain and the appropriate tools (Fourier Transform, Laplace Transform, convolution)  and moves to the discrete signal domain and the appropriate tools (Dicrete Fourier Transform, Z-Transform, linear and cyclic convolution). The next topic is Filter Design and Implementation and Adaptive Filtering (wiener filtering). After a brief refresh of statistics and probability theory it covers all the basic machine learning theory on supervised and unsupervised learning and leads to current advances in deep learning.

Prof. Nikolaos Mitianoudis (nmitiano@ee.duth.gr)

Assist. Prof. Ilias Theodorakopoulos (Iltheodo@ee.duth.gr)

 

Course Structure

1. Analog Signals. Continuous Fourier Transform
2. Sampling. Nyquist Theorem.
3. The Z-Transform. System Transfer Function.
4. Discrete Fourier Transform, Fast Fourier Transform.
5. Filter Design. IIR and FIR Filter Design
6. Basic Probabillity Theory
7. Introduction to Adaptive Filtering - Wiener Filters
8. Introduction to Machine Learning. Classification vs Regression
9. Linear and Non-linear regression
10. Unsupervised Clustering: K-Means, ISODATA
11. Supervised Clustering: K-NN, Support Vector Machines
12. Introduction to Deep Learning and Modern Deep Learning Architectures

Recommended Textbooks

1. A. V. Oppenheim, R. Schaffer, Digital Signal Processing
2. S. Mitra, Digital Signal Processing, 4th edition, McGraw-Hill, 2010.
3. M. Hayes, Statistical Digital Signal Processing and Modeling
4. S. Theodoridis, Pattern Recognition.
5. R. O. Duda, D. G. Stork, P. E. Hart, Pattern Classification
6. C. M. Bishop, Pattern Recognition and Machine Learning

Ημερομηνία δημιουργίας

Παρασκευή 10 Νοεμβρίου 2023