LFU:online

Students gain a basic understanding of data-driven environmental modelling. By the end of the course, they are able to: (1) assess the contributions, strengths, limitations, and major trends of traditional machine learning (ML) methods in geoscience applications; (2) understand and compare the epistemic modelling cultures of physically-based geosciences and data-driven ML; (3) work with quantitative geoscientific data and identify their key characteristics; and (4) apply traditional ML algorithms and Bayesian inference methods to geoscientific problems and selected case studies.

Machine learning (ML) has rapidly become a key tool in geoscientific research, bridging traditionally separate research communities. This VU introduces data-driven environmental modelling at the intersection of traditional ML (excluding deep learning) and Bayesian statistics. After a short introduction to geoscientific practice, ML, probability theory, and statistics, the course focuses on hands-on applications of classification, clustering, feature extraction, Bayesian inference, and data fusion using Kalman filtering. Examples are drawn from glaciers, mountain permafrost, mountain meteorology, and natural hazards.

Short mid-term exam (20%) and a written report on a solved problem (80%). Active in-class participation is expected.

The course is aimed at MSc-level students in computer science, engineering, and geosciences. Deep prior knowledge of geosciences or machine learning is not required but is advantageous. Students are expected to have prior exposure to linear algebra, probability, and statistics, as well as coding experience in Python. Priority is given to MSc students; however, subject to capacity, advanced BSc students may also enroll.

Allocation of places in courses with a limited number of participants (PS, SE, VU, PJ)

In courses with a limited number of participants, course places are allocated as follows:
1. Students for whom the study duration would be extended due to the postponement are to be given priority.
2. If the criteria in no. 1 do not suffice, first, students for whom this course is part of a compulsory module are to be given priority, and second, students for whom this course is part of an elective module.
3. If the criteria in no. 1 and 2 do not suffice, the available places are drawn by random.

Curriculum BA Computer Science 2019

Curriculum MA Computer Science 2021