Seminar MedientechnikHauptseminar Medientechnik

Media Technology Seminar (Winter Semester 2017/18)

Lecturer (assistant) Eckehard Steinbach, Christoph Bachhuber, Dmytro Bobkov, Matti Strese, Jingyi Xu, Alexandra Zayets
Audience see TUMonline
Teaching language German
Semester weekly hours
Scheduled dates see TUMonline
Registration see TUMonline


Selected topics in media technology. The focus is on current research topics and new technologies. The participants study recent publications, prepare a summary in the form of a scientific paper and present the topic to the audience.


The following literature is recommended:
- will be announced in the seminar


rendered: 2018-02-25 00:33:12


Thema WS17/18: DEEP LEARNING- Theoretische Grundlagen und die Anwendungen in der Medientechnologie

Die geplanten Vortragsthemen für die Teilnehmer sind:

  1.  Basics of  Information Theory and Machine Learning 
  2.  Deep Feedforward Networks 
  3.  Regularization for Deep Learning 
  4.  Optimization for Training Deep Models 
  5.  Convolutional Networks
  6.  Deep Generative Models  
  7.  Monte Carlo Methods
  8.   Representation Learning
  9.  Understanding Neural Networks through Visualization
  10.  Autoencoders 
  11.  Structured Probabilistic Models for Deep Learning
  12.  Approximate Inference

 Eine detailliertere Beschreibung (inklusive Literatur zum Start) finden Sie hier.


Der Kickoff-Termin findet statt am: 18.10.17, 13:15-14:45 Uhr, Raum 0406.