| Detailed information |
| Original study plan |
Master's programme Computer Science 2021S |
| Objectives |
Participants will obtain an overview of standard methods in the fields of pattern classification, machine learning, and statistical data modelling. The focus of the class will very much be on application-oriented issues: students will learn the basic concepts and methods in the field, and understand what is involved when applying these methods to complex classification and recognition problems. To this end, the accompanying exercise class (UE) will permit student to experiment with learning and classification algorithms in a complex, non-trivial real-world application problem.
|
| Subject |
Bayes classification and Bayes error; density estimation; nearest-neighbour classification; standard classifiers in machine learning (decision trees, Naive Bayes, feedforward neural networks, support vector machines, ensemble methods); empirical evaluation of classifiers; clustering and mixture models; dimensionality reduction and data projection methods; Markov processes and Hidden Markov Models.
|
| Criteria for evaluation |
Written exam at the end of the semester.
|
| Methods |
Standard lecture series, with class materials (lecture slides) regularly provided in electronic form.
|
| Language |
English |
| Study material |
Lecture slides will regularly provided in electronic form.
No further materials required.
|
| Changing subject? |
No |
| Corresponding lecture |
(*)in collaboration with 921PECOMLPU20: UE Machine Learning and Pattern Classification (1.5 ECTS) equivalent to
921PECOMLPK13: KV Machine Learning and Pattern Classification (4.5 ECTS)
|
