Moses - Deep Learning for Communications

Display language

To modulepage Generate PDF

Diese Modulbeschreibung gilt nicht im aktuellen Semester.

Die Modulbeschreibung gilt nur für den Zeitraum SoSe 2020 - SoSe 2022.

Module / Version:
#40979 / #1

Validity:
SoSe 2020 - SoSe 2022

Default display language:
English

Module title:
Deep Learning for Communications

Credits:
3

Responsible person:
Schaefer, Rafael

Grading:
benotet

Type of exam:
Portfolioprüfung

Zugehörigkeit

Faculty:
Fakultät IV

Institute:
Institut für Telekommunikationssysteme

Area of expertise:
34331900 FG Informationstheorie und deren Anwendungen

Prüfungsausschuss:
No information

Kontakt

Office:
HFT 6

Contact person:
Schaefer, Rafael

E-mail address:
rafael.schaefer@tu-berlin.de

Website:

http://www.commit.tu-berlin.de/menue/lehre/parameter/en/

Learning Outcomes

After completing this module the students will have a solid understanding of the principles of deep learning and its application in communications. They will get familiar with the current literature in this emerging field and will strengthen their skills to understand, discuss, and present scientific contributions.

Content

Artificial intelligence and machine learning are experiencing a considerable interest these days and their applications now extend into almost every industry and research domain. Particularly deep learning has led to many recent breakthroughs in various domains including computer vision, speech recognition, and natural language processing. This aim of this seminar is to provide an introduction into the concept of deep learning and to present and discuss its application in communications via student presentations of scientific research papers. These include topics such as neural network-based communication systems, channel modeling via generative adversarial networks, code-design via autoencoders, and many others.

Module Components

Pflichtgruppe:

All Courses are mandatory.

Course Name	Type	Number	Cycle	Language	SWS	VZ
Deep Learning for Communications	SEM	34331900 L 005	SoSe	English	2

Workload and Credit Points

Deep Learning for Communications (SEM):

Workload description	Multiplier	Hours	Total
Attendance	15.0	2.0h	30.0h
Pre/post processing	15.0	4.0h	60.0h
			90.0h(~3 LP)

The Workload of the module sums up to 90.0 Hours. Therefore the module contains 3 Credits.

Description of Teaching and Learning Methods

This seminar starts with a small recap and introduction into the topic of deep learning for communications by conventional frontal teaching in class using slides and blackboard. Subsequently, each week a scientific paper is presented by a student and discussed.

Requirements for participation and examination

Desirable prerequisites for participation in the courses:

Prerequisite for participation to courses are a mathematical background at the level of beginning MS students in Electrical Engineering. A background in deep learning/machine learning is desirable, but a brief recap will be given at the beginning.

Mandatory requirements for the module test application:

This module has no requirements.

Module completion

Grading

graded

Type of exam

Portfolio examination

Type of portfolio examination

100 Punkte insgesamt

Language

German/English

Test elements

Name	Points	Categorie	Duration/Extent
(Deliverable assessment) Presentation	66	oral	45min
(Deliverable assessment) Mini-report	34	written	4-6 pages

Grading scale

Notenschlüssel »Notenschlüssel 3: Fak IV (3)«

Gesamtpunktzahl	1.0	1.3	1.7	2.0	2.3	2.7	3.0	3.3	3.7	4.0
100.0pt	85.0pt	80.0pt	75.0pt	70.0pt	65.0pt	60.0pt	55.0pt	50.0pt	45.0pt	40.0pt

Test description (Module completion)

The exam consists of a presentation of a scientific paper followed by a written mini-report on the topic.

Duration of the Module

The following number of semesters is estimated for taking and completing the module:
1 Semester.

This module may be commenced in the following semesters:
Sommersemester.

Maximum Number of Participants

The maximum capacity of students is 15.

Registration Procedures

Course teaching and organization is supported by an ISIS course. Module examination enrollment is done via QISPOS. Details will be given in the first meeting.

Recommended literature
T. J. O’Shea and J. Hoydis, "An introduction to deep learning for the physical layer," IEEE Transactions on Cognitive Communications and Networking, vol. 3, no. 4, pp. 563-575, December 2017
I. Goodfellow, Y. Bengio and A. Courville, "Deep Learning" MIT Press, 2017

Assigned Degree Programs

Semester:

Zeige auch ausgelaufene Studiengänge und StuPOs

This module is used in the following Degree Programs (new System):

	Studiengang / StuPO	StuPOs	Verwendungen	Erste Verwendung	Letzte Verwendung
This module is not used in any degree program.

Miscellaneous

No information

Deep Learning for Communications