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

Lehrveranstaltungen:

• • Machine Learning in Signal Processing
    (Vorlesung, 3 SWS, Veniamin Morgenshtern, Mo, 12:15 - 13:45, HB; Fr, 10:15 - 11:45, HC; Blockveranstaltung 9.4.2018-16.4.2018 Mo, 12:15 - 13:45, 05.025; Blockveranstaltung 13.4.2018-20.4.2018 Fr, 10:15 - 11:45, 05.025; ab 23.4.2018; The lecture takes place in the lecture halls HB and HC from April 23 on.)
  • Supplements for Machine Learning in Signal Processing
    (Übung, 1 SWS, Veniamin Morgenshtern, Di, 14:15 - 15:45, 05.025)

Inhalt:

This course is an introduction into modern statistical machine learning and artificial intelligence. The goal is to familiarize the students with import machine learning algorithms and explain their design principles. First we will study methods that rely on mathematical modeling to find a good representation for the signal class (features). The topics in this section include: linear regression, features and efficient signal representations (splines, wavelets, short time Fourier transform, Gabor frames, HOG and SIFT features for images), regularization, short introduction to convex optimization, logistic regression, bias-variance tradeoff, cross-validation, learning in high-dimensional spaces vs. learning in low dimensional spaces, support vector machines. Then we will study methods to learn good features from data automatically. The topics in this section include: dictionary learning and deep neural networks. We will also discuss unsupervised learning algorithms: principle component analysis, k-means clustering, nearest neighbors, mixture of Gaussians, the EM algorithm. We will cover many examples: denoising and compression of sound and images, compressive sensing, digit classification in the MNIST dataset, spam filters. Time permitting, we will discuss regression trees and boosting.

Lernziele und Kompetenzen:

Students:
Learn how to solve practical problems using machine learning algorithms.
Learn how to control the quality of the learning procedure.
Learn how to debug machine learning algorithms.
Understand theoretical aspects that underpin the design of new algorithms.
Understand the importance of statistics and optimization in machine learning.
Learn classical efficient signal representations.
Understand how deep neural networks are able to learn very efficient signal representations automatically.
Learn standard packages for doing machine learning in Python: numpy, scikit-learn, scikit-image, cvxpy.

Literatur:

Literature:
T. Hastie, R. Tibshirani, J. Friedman: The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition.
S. Mallat: A Wavelet Tour of Signal Processing, Third Edition: The Sparse Way.
I. Goodfellow, Y. Bengio, A. Courville: Deep learning.

Weitere Informationen:

Verwendbarkeit des Moduls / Einpassung in den Musterstudienplan:
Das Modul ist im Kontext der folgenden Studienfächer/Vertiefungsrichtungen verwendbar:

1. Advanced Signal Processing & Communications Engineering (Master of Science)
   (Po-Vers. 2016w | TechFak | Advanced Signal Processing & Communications Engineering (Master of Science) | Masterprüfung | Pflichtmodule | Machine Learning in Signal Processing)
2. Communications and Multimedia Engineering (Master of Science)
   (Po-Vers. 2011 | TechFak | Communications and Multimedia Engineering (Master of Science) | Masterprüfung | Wahlpflichtmodule | Technische Wahlpflichtmodule)
3. Communications and Multimedia Engineering (Master of Science)
   (Po-Vers. 2011 | TechFak | Communications and Multimedia Engineering (Master of Science) | Masterprüfung | Wahlmodule | Technische Wahlmodule)
4. Informations- und Kommunikationstechnik (Master of Science)
   (Po-Vers. 2016s | TechFak | Informations- und Kommunikationstechnik (Master of Science) | Schwerpunkte im Masterstudium | Schwerpunkt Eingebettete Systeme | Wahlpflichtmodule | Wahlpflichtmodul aus EEI im Schwerpunkt Eingebettete Systeme)
5. Informations- und Kommunikationstechnik (Master of Science)
   (Po-Vers. 2016s | TechFak | Informations- und Kommunikationstechnik (Master of Science) | Schwerpunkte im Masterstudium | Schwerpunkt Kommunikationsnetze und Übertragungstechnik | Wahlpflichtmodule | Wahlpflichtmodul aus EEI im Schwerpunkt Kommunikationsnetze)
6. Informations- und Kommunikationstechnik (Master of Science)
   (Po-Vers. 2016s | TechFak | Informations- und Kommunikationstechnik (Master of Science) | Schwerpunkte im Masterstudium | Schwerpunkt Multimediasysteme | Wahlpflichtmodule | Wahlpflichtmodul aus EEI im Schwerpunkt Multimediasysteme)

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

1. Prüfer:

Veniamin Morgenshtern

Termin: 13.07.2018, 10:15 Uhr, Ort: HC Bio
Termin: 17.04.2019, 11:00 Uhr, Ort: H 8 TechF