Seminar Deep Learning

Summer term 2017

News

The registration for the seminar presentation is now possible in the examination office. Be aware of the opening hours. Registration is open until 19th of July.
The List of Paper Titles and Partitipants working on them is now available

Overview

General
Dates and Rooms
Additional material
Links

General

On this page, you find information on the scientific seminar "Deep Learning", which will be held during summer term 2017 by Christoph Doell at the Otto-von-Guericke University of Magdeburg.

Learning content of the seminar

The seminars purpose is to build and train key competences scientific working, research, scientific writing, presenting of a scientific topic as well as content knowledge on the topic of Deep Learning.

The student activities consist of several parts: After the preparation of a scientific topic, students write and submit a research paper on it. At the end of the course the topic will be presented to all participants.

Technical content of the seminar

The technical content of the seminar are recent topics from the area of neural networks especially from deep learning. As this is a scientific seminar and we focus on recent developments, we assume that you have a general knowledge of neural networks already. Otherwise, we do not recommend to attend the seminar as we focus on current research topics. If you are interested in an introduction to neural networks feel free to participate in the neural networks lecture. Possible seminar topics will be announced here later.

Language

The papers will be given in english. Nevertheless you may choose english or german as language for your paper and your presentation. Still, there are two reasons to choose english: 1) It can be assumed that other participants (who do not speak german) want to ask questions or participate in discussions. 2) Practicing english is probably helpful when writing your thesis and might also be benefitial for your future career.

Course

At the first appointment - 03.04.2017, 4th of April 2017 9 - 11 a.m in G29-301, all available topics will be shortly presented to the participants. Each participant gets his own topic and correspondingly related literature. In the following weeks you write a scientific paper which summarizes your topic and explains the state of the art. Those are submitted until 30th of July (30.07.2017) and rated. In the following week 2017-08-07 (07.08.2017) to 2017-08-09 (09.08.2017) the block seminar part will take place, where all participants present their work. Presentations last 20 minutes, followed by 5 to 10 minutes of discussion. As the number of participants is not fixed yet, the time slots for the presentations will be announced later.

Required knowledge

Basics in Computer Science
Neural Networks
(helpful) Machine Learning

Organizational Rules for participation

The seminar is planned for up to 20 students in computer science and other related fields.

All participants are asked to register via FIN Registration Service.

The number of students interested in this course is much higher than we expected. This is why we would like to ask everyone of you to send us a short "application" letter (approx 250 words) until April 1st 2017, in which you explain what motivates you to participate in this course. Please let us also know if you want to take one or two topics to work on (3CP or 6CP). From all received letters we will pick up to 20 students to take part in this seminar.

For participation, the following efforts are needed:

successful oral presentation
6-8 pages of an own research paper
Participate in at least 90% of the presentations

For participating in the scientific seminar, we give 3 credit points. You can gain 6CP, by doubling the efforts e.g. writing two papers, holding two presentations, etc.

Appointments and rooms

	Weekday	Time	Room	Start
Seminar	monday	09.15-10.45	G29-301	03.04.2017

Important dates:

Announcement of topics: 03.04.2017 (2017-04-03)
Paper submission: 30.07.2017 (2017-07-30)
Paper Presentation: 07.08.2017 to 09.08.2017

Additional material

Participants and Papers
Latex2e Proceedings Teamplate for paper: LNCS-Style (Please use LaTeX!)
Slides on the topic How-to-present
Website Readability Score
Youtube Channel Deep Learning TV
Online-Book Deep Learning Book (Goodfellow 2016)