Digital Modelling (3cr)

Objective

Students will be competent in using the mathematical methods described in the course contents to solve practical mathematical problems.

Content

Basics of machine learning:
Minimization by gradient descent
Linear regression
Logistic regression
Neural networks

Location and time

The schedules can be found in the timetable at https://lukkarikone.seamk.fi/. Timetables are published for the next six weeks. The first six weeks of autumn are published by Midsummer and the first six weeks of spring by Christmas. Timetables may be subject to changes.

Materials

To be announced at the beginning of the course.

Teaching methods

The study involves contact teaching and requires attendance at the SEAMK campus.

Student workload

The workload of the study is designed so that one credit corresponds to an average of 27 hours of student work to achieve the learning objectives. The actual time required varies individually, e.g., due to prior knowledge.

Assessment criteria, satisfactory (1)

satisfactory (1-2): The student knows and understands to a satisfactory extent the basic concepts and methods discussed in the course, and is able to apply them to usual problems.

Assessment criteria, good (3)

good (3-4): The student is familiar with the concepts and methods discussed in the course, and is able to apply them to different types of problems. The student is able to combine the accumulated knowledge and skills with previous experiences in the subject.

Assessment criteria, excellent (5)

excellent (5): The student is familiar with the concepts and methods discussed in the course, and is able to apply them to a variety of different problems. The student has demonstrated creativity and innovation, and is able to find new meanings when applying what they have learned

Qualifications

Algebra and geometry, Vectors and matrices, Differential and integral calculus, Automation technology mathematics

Further information

80% attendance requirement at classes.