Pattern Recognition and Deep Learning

Year:
1st year/2nd year
Semester:
S1
Programme main editor:
Onsite in:
Remote:
ECTS range:
6 ETCS

Professors

img
Friedhelm Schwenker
UULM

Prerequisites:

Pedagogical objectives:

Students acquire knowledge about different methods and algorithms of pattern recognition and deep artificial neural networks. In exercises, students are able to implement the basic algorithms, will apply pattern recognition principles to technical applications, and learn how to evaluate the performance of classifiers.

Evaluation modalities:

Oral exam

Description:

Topics include:

In this course the basic topics on statistical pattern recognition and deep neural networks are introduced:

  • Introduction to statistical and neural pattern recognition
  • Linear and nonlinear classifiers
  • Kernel methods and learning deep neural network
  • Feature extraction, selection, and reduction
  • Applications and system performance evaluation

Required teaching material

Literature: • Bishop, Chris: Pattern Recognition and Machine Learning, Springer, 2007 • Theodoridis, Sergios & Koutroubas, Konstantionos, Pattern Recognition, Academic Press, 2010 • Charu Aggarwal: Neural Networks and Deep Learning, Springer, 2018 • A.E. Bryson, Y.-C. Ho: Applied Optimal Control, Hemisphere Publishing Corporation, 1975 • Moodle Course at https://elearning.saps.uni-ulm.de/ - Account needed at SAPS of UUlm

Teaching volume:
lessons:
20 hours
Exercices:
16 hours
Supervised lab:
Project:

Devices:

  • Laboratory-Based Course Structure
  • Open-Source Software Requirements