Page tree

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

9. Course learning outcomes

-Description of the learning outcomes provided The learning outcomes provided to students by the course
- See the competence map (https://flamma.helsinki.fi/content/res/pri/HY350274).Obtains deeper knowledge of domain skills in machine learning: Can describe

  1. Explain the basic formulation of machine learning as minimising the expected risk, and

...

  1. identify alternative formulations for the risk.

...

  1. Explain the core tasks of unsupervised and supervised learning, and identify also more advanced learning setups.
  2. Derive practical loss functions starting from the formal definition, and

...

  1. explain the relationship between probabilistic models and loss minimisation.

...

  1. Derive algorithms suitable for unsupervised learning tasks and (regularised) linear methods for classification and regression.
  2. Implement algorithms suitable for unsupervised learning tasks and (regularised) linear methods for classification and regression

...

  1. .
  2. Implement some non-linear classification methods such as random forests and support vector machines.

10. Course completion methods

The course is completed via by completing a combination set of exam and exercises, and both parts need to be passed to complete the course. Part . Significant part of the exercises involve programming.Completing the course with separate exam requires solving a small research project.

11. Prerequisites

Introduction to Machine Learning or equivalent knowledge. Programming skills, especially with python, are highly recommended. Basic knowledge of linear algebra, statistics, and analysis, especially differentiation, is required.

12. Recommended optional studies

...

Formulation of machine learning as risk minimisation and as probabilistic modellingproblems. Different kinds of machine learning tasks, covering also advanced setups such as transfer learning. Common optimisation approaches for machine learning. Unsupervised learning methods: clustering, factor analysis, matrix factorisation, non-linear dimensionality reduction. Supervised learning methods: Linear and non-linear classifiers, kernel methods, decision trees and forests, boosting. 

...

-What kind of literature and other materials are read during the course (reading list)?

Course The course material are slides, supported by the course book:

Kevin P. Murphy "Machine Learning: A Probabilistic Perspective", MIT Press, 2012.

The course book material is complemented with additional publicly available material, and the course book may change in future.-Which works are set reading and which are recommended as supplementary reading?The course planned for 2017 covers Chapters 1-14, 16, 19, 25, 27-18 of the course book.

15. Activities and teaching methods in support of learning

...

The primary mode of instruction consists of lectures and exercise sessions with active guidance, supported by other forms of teaching methods when applicable. The students are encouraged to attend the lectures and they need to solve exercise problems including problems involving programming tasks to reach the learning outcomes related to implementation skills. Some of the exercise problems are formulated in an open manner to support acquisition of problem-solving skills, and require written presentation to facilitate learning of scientific presentation skills. the exercise sessions. Support channel (online chat) is provided for students.

16. Assessment practices and criteria, grading scale

Grading scale is 1...5.

The grading is based on a combination of a course exam and exercises. One should obtain half of the points for both the exam and the exercises to pass the course.The course can alternatively be taken by completing a separate exam and a project work. One should obtain half of the points for both the exam and the project to pass the coursethe points received from the exercises.