SGN-13006 Introduction to Pattern Recognition and Machine Learning, 5 cr
Additional information
Lectures are in English, but exercise sessions are given in both Finnish and English.
Person responsible
Joni Kämäräinen
Lessons
| Implementation | Period | Person responsible | Requirements |
| SGN-13006 2019-01 | 1 |
Joni Kämäräinen |
Active participation, homeworks, exercises and exam. |
Learning Outcomes
This course will provide a broad introduction to Pattern Recognition (PR) and Machine Learning (ML). The course is programming oriented concentrating on models of learning (data structures to establish these models) and methods of learning (algorithms to modify data structures according to training data). After the course students will know the main approaches to machine learning starting from early ideas to the most recent ones. Students will also obtain skills to implement ML&PR methods and evaluate them with real data.
Content
| Content | Core content | Complementary knowledge | Specialist knowledge |
| 1. | Basic work flow in pattern recognition and machine learning. Linear models of regression and classification as the starting point. | ||
| 2. | Concept learning. | ||
| 3. | Decision tree learning | Random forests | |
| 4. | Bayesian learning and probability density estimation | ||
| 5. | Prolog language and the principal idea of inductive logic programming. | ||
| 6. | Multi-layer perception neural networks and support vector machines. | ||
| 7. | Unsupervised learning including clustering, self-organising map and linear methods (principal component analysis) | ||
| 8. | Pattern recognition and machine learning in robotics and re-inforcement learning. |
Instructions for students on how to achieve the learning outcomes
You must actively participate the lectures and do the exercises. In particular, familiarize yourself with the exercise questions before the exercise session.
Assessment scale:
Numerical evaluation scale (0-5)
Partial passing:
Study material
| Type | Name | Author | ISBN | URL | Additional information | Examination material |
| Book | Elements of Statistical Learning, 2nd edition | Trevor Hastie, Robert Tibshirani, Jerome Friedman | Covers all the required methods, but is rather statistical approach. Mainly the random forest part is taken from this book. | Yes | ||
| Book | Machine Learning | Tom Mitchell | 0070428077 | Contents of many lectures follow this book | Yes | |
| Book | Statistical Pattern Recognition, 3rd Edition | Andrew R. Webb, Keith D. Copsey | 978-0-470-68227-2 | Very good book about the topic from practioners. Mainly the support vector machines part is taken from this book. | Yes |
Additional information about prerequisites
No mandatory requirements, but students are assumed to have good knowledge about BSc level engineering mathematics and programming.
Correspondence of content
| Course | Corresponds course | Description |
| SGN-13006 Introduction to Pattern Recognition and Machine Learning, 5 cr | SGN-2506 Introduction to Pattern Recognition, 4 cr | |
| SGN-13006 Introduction to Pattern Recognition and Machine Learning, 5 cr | SGN-13000 Introduction to Pattern Recognition and Machine Learning, 5 cr |