Pattern Recognition and Machine Learning R1a

Overview

Credits points: 6

Workload:
60 hours course attendance; 120 hours self-study

Semester:
Semester: winter

Language: English

Module type: elective

Module usability: M.Sc. Electrical Communication Engineering, M.Sc. Elektrotechnik

Module duration: one semester

Required qualifications:
Knowledge of some contents from mathematics lectures (stochastics or discrete
structures, analysis, linear Algeba) or comparable knowledge and skills

Competences to be acquired

Knowledge: theoretical basics of pattern recognition (probabilistic point of view)

Ability to use of parameter estimation techniques for different models

Development of new models

Evaluation of practical applications and independent development of new applications

Courses

Content

  • Fundamentals (e.g. stochastics, model selection, curse of dimensionality, decision and information theory), distributions (e.g. multinomial, dirichlet, Gaussian and student distribution, nonparametric estimation of distributions)
  • Linear models for regression, linear models for classification
  • Kernel functions and advanced neural networks (e.g. vonvolutional neural networks, radial basis function networks), Gaussian processes

Learning outcomes

  • Understanding the theoretical basics of pattern recognition and machine learning
  • Learning about parameter estimation techniques
  • Ability to develope of new models

Details

  • Lecturer: Bernhard Sick and team
  • Teaching method: lecture and exercises
  • SWS: 4
  • Credit points: 6
  • Examination: oral exam (30 min)
  • Course identifier: FB16-6973