| NUMMER: | 212101 |
| KÜRZEL: | semToDLSProc |
| MODULBEAUFTRAGTE:R: | Dr. Anand Subramoney |
| DOZENT:IN: | Dr.Anand Subramoney |
| FAKULTÄT: | Fakultät für Informatik |
| SPRACHE: | English |
| SWS: | 2 |
| CREDITS: | 3 |
| ANGEBOTEN IM: | jedes Wintersemester |
LINK ZUM VORLESUNGSVERZEICHNIS
Hier entlang.
LERNFORM
Vertiefungsseminar
LERNZIELE
After the successful completion of this course the students∙ understand models for sequence processing including recurrent networks and transformers,
∙ understand core challenges of training these models working with time, and
∙ will be able to contextualize current publication with respect to those challenges.
INHALT
Natural language processing, robotics, video processing, stock market forecasting and othersimilar tasks require models that can deal with sequence data and understand temporal dependencies. Two major classes of models that have been designed to deal with sequence data
are recurrent neural networks (RNNs/LSTMs) and transformer architectures. Designing and
understanding these models is a very active and diverse area of research. Applications of these
models are also widespread. The recent explosion of interest in topics such as language modelling and machine translation is based on advances in these models which includes GPT-3,
DALL-E, etc.
In this course you’ll first understand the fundamentals of recurrent neural networks and
transformers that led to these breakthroughs. Then we’ll go through and discuss both seminal
and recent research papers on these topics to throw light on algorithms and challenges in this
field.
VORAUSSETZUNGEN CREDITS
a) Successful presentation and positively evaluated written elaborationb) Semester-long; successfully completed practical assignment(s) or practical project
EMPFOHLENE VORKENNTNISSE
Keine
AKTUELLE INFORMATIONEN
Stellenwert der Note für die Endnote: 3 / 112