Machine Learning in Signal Processing (MLSIP)5 ECTS
(englische Bezeichnung: Machine Learning in Signal Processing)
(Prüfungsordnungsmodul: Technische Wahlmodule)

Lehrveranstaltungen:

• • Machine Learning in Signal Processing
    (Vorlesung, 3 SWS, Veniamin Morgenshtern, Mi, 8:15 - 9:45, H5; Fr, 12:15 - 13:45, H5)
  • Supplements for Machine Learning in Signal Processing
    (Übung, 1 SWS, Veniamin Morgenshtern, Do, 12:15 - 13:45, 05.025)

Inhalt:

This course is an introduction into statistical machine learning and artificial intelligence. The special emphasis is on applications to modern signal processing problems. The course is focused on design principles of machine learning algorithms. First we will study basic methods for regression and classification: linear regression, logistic regression, the nearest neighbors algorithm. Based on these examples, we will discuss the fundamental trade-off between the flexibility of the model and the ability to fit the model based on the moderate amount of training data. We will contrast learning in high-dimensional spaces vs. learning in low dimensional spaces. Next, we will study methods that help make linear models flexible: polynomial features and splines. When these tools are used, regularization is crucial. Next, we will discuss structured signal representations: short-time Fourier transform, Gabor frames, wavelets. We will focus on the importance of sparsity in signal representations. This will lead us to compressed sensing and to other modern convex-optimization-based methods for signal denoising, reconstruction, and compression. We will review key concepts in convex optimization, study the LASSO, support vector machines, the idea of kernels. The last part of the course will focus on the breakthrough new technologies for computer vision: HOG and SIFT features for image recognition and the deep learning. Time permitting we will discuss unsupervised learning algorithms: principle component analysis, k-means clustering, mixture of Gaussians, the EM algorithm.
The course contains exercises: 30% mathematical and 70% programming in Python. You will be asked to implement basic machine learning and signal processing algorithms yourself. For more advanced algorithms, you will practice using powerful numerical and optimization libraries (numpy, cvxpy, scikit-learn, pywavelets, pytorch).

Lernziele und Kompetenzen:

Students are able to:

• Apply standard machine learning and signal processing algorithms to design solutions to practical problems in new domains.

• Use standard packages for machine learning in Python: numpy, cvxpy, scikit-learn, pywavelets, pytorch.

• Choose appropriate algorithms and signal representations for the problem at hand.

• Debug and calibrate machine learning algorithms. Develop simple modification to the standard algorithms as appropriate to the problem at hand.

• Rapidly discover, understand, and apply advanced algorithms and signal representations that were not covered in class.

• Explain the theoretical aspects that underpin the design of new algorithms.

• Explain the importance of statistics and optimization in machine learning.
  

Literatur:

Literature:

• T. Hastie, R. Tibshirani, J. Friedman: The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition, Chapters 1–7.
• Ng: Lecture notes and materials for Stanford CS229 class. Lecture Notes and Exercises.
• M. Kon: Lecture notes on basics of wavelets.
• M. Nielsen: Neural networks and deep learning.

Weitere Informationen:

Verwendbarkeit des Moduls / Einpassung in den Musterstudienplan:

1. Communications and Multimedia Engineering (Master of Science)
   (Po-Vers. 2011 | TechFak | Communications and Multimedia Engineering (Master of Science) | Masterprüfung | Wahlmodule | Technische Wahlmodule)

Studien-/Prüfungsleistungen:

Machine Learning in Signal Processing (Prüfungsnummer: 84401)

Prüfungsleistung, Klausur, Dauer (in Minuten): 90, benotet, 5 ECTS

Anteil an der Berechnung der Modulnote: 100.0 %

Prüfungssprache: Englisch

Erstablegung: WS 2018/2019, 1. Wdh.: SS 2019

1. Prüfer:

Veniamin Morgenshtern