Seminar Machine Learning (SE)
Course Description
The aim of machine learning is to develop methods that are able to automatically detect pattern in data and to use these to predict future outcomes of interest. Hence, machine learning and statistics are closely related fields, yet differing in terminology and emphasis. This seminar aims to give an introduction to the field with basic concepts and algorithms as well as examples drawn from different application domains. Topics include, amongst others, classification, boosting, graphical models, approximate inference, neural networks and deep learning.
Registration in the respective Moodle course is obligatory.
Prerequisites
Participants should be familiar with basic concepts of probability, multivariate calculus, linear algebra and computer programming.
Also, PhD students with related fields of research are welcome.
Course Learning Objectives
Broad overview on topics of machine learning.
Course Structure
Block seminar.
Literature and Sources
-
Pattern Recognition and Machine Learning, Chris Bishop, Springer 2006.
-
Machine Learning: a Probabilistic Perspective, Kevin Murphy, 2012
-
Deep learning book, Goodfellow et al, 2016
-
The elements of statistical learning, Trevor Hastie, Robert Tibshirani and Jerome Friedman, Springer 2001.
-
Information theory, inference and learning algorithms, David Mackay, CUP 2003
www.quantlet.de (source codes)