Lecture: Foundations of Machine Learning and Data Mining
Lecturers: Ralf Möller
Prerequisites:
Basic knowledge in Computer Science, Discrete Mathematics, Linear Algebra, Mathematical Analysis, Probability Theory, Statistics
Knowledge in Intelligent Autonomous Agents would be a plus
Educational Objectives:
After the course Bachelor students have expertise about basic machine learning and data mining techniques as well as analytical skills to match pros and cons of techniques to requirements in applications. With this course students develop prerequisite capabilities for a more in-depth Master-level course on Cognitive Robotics or Probabilistic Models of Human and Machine Intelligence.
Content :
- Introduction
- Decision trees
- Decision trees for real-world scenarios
- Incremental learning: Version spaces
- Uncertainty
- Bayesian networks
- Learning parameters of Bayesian networks
BME, MAP, ML, EM algorithm - Learning structures of Bayesian networks
- kNN classifier, neural network classifier, support vector machine (SVM) classifier
- Clustering
Distance measures, k-means clustering, nearest neighbor clustering - Ensemble Learning
- Reinforcement Learning
- Computational Learning Theory
Acknowledgments:
Slides are taken from courses by Ethem Alpaydin, Stuart Russell, Hwee Tou Ng, Y. Hou, Eamonn Keogh, Xiaoli Fern, Carla P. Gomes, Nathalie Japkowicz et al.Literature:
- Artificial Intelligence: A Modern Approach (Third Edition), Stuart Russel, Peter Norvig, Prentice Hall, 2010
Chapters 13, 14, 18, 20, 21. - Data Mining: Practical Machine Learning Tools and Techniques Mark Hall, Ian Witten and Eibe Frank , Morgan Kaufmann, 2011
Previous exams
Ralf Möller