Disciplines/fields: Machine Learning

Duration: 4 sessions

Course Content

The course will cover fundamental concepts underlying machine learning from given data and popular learning algorithms in the field of supervised and unsupervised learning. We will start with an example problem and discuss when and how to learn in a simple example setting. Then we will introduce different fundamental learning models and learning paradigms such as parametric models, lazy learning, Bayes learning, decision trees, and their role in an exemplary learning pipeline. The latter will also highlight important aspects such as preprocessing, model selection, and model regularization. Afterwards, we have a look at more complex supervised learning schemes, in particular support vector machines, neural networks, and Gaussian processes. Finally, we will discuss unsupervised learning methods, in particular clustering and dimensionality reduction techniques for efficient data analysis.

Objectives

  • Conceptual: To understand when and why machine learning from data can work, and what are the fundamental steps, which make it successful.
  • Methodological: Learn the basics of fundamental machine learning models, and when and how to apply those for given data.

Literature

Pattern Recognition and Machine Learning, Chris Bishop, Springer, 2006
Data Classification, Charu Aggarwal, CRC, 2014
Learning from Data, Yaser Abu-Mostafa, AML, 2012
Machine Learning, Kevin Murphy, MIT, 2012

Vita

Barbara Hammer received her Ph.D. in Computer Science in 1995 and her venia legendi in Computer Science in 2003, both from the University of Osnabrück, Germany. From 2000-2004, she was chair of the junior research group 'Learning with Neural Methods on Structured Data' at University of Osnabrück before accepting an offer as professor for Theoretical Computer Science at Clausthal University of Technology, Germany, in 2004. Since 2010, she is holding a professorship for Theoretical Computer Science for Cognitive Systems at the CITEC cluster of excellence at Bielefeld University, Germany. Several research stays have taken her to Italy, U.K., India, France, the Netherlands, and the U.S.A. Her areas of expertise include hybrid systems, self-organizing maps, clustering, and recurrent networks as well as applications in bioinformatics, industrial process monitoring, or cognitive science. She has been chairing the IEEE CIS Technical Committee on Data Mining in 2013/2014, and she is chair of the Fachgruppe Neural Networks of the GI and vice-chair of the GNNs. She has published more than 200 contributions to international conferences / journals, and she is coauthor/editor of four books.

Link

http://www.techfak.uni-bielefeld.de/~bhammer/