Linear Algebra and Learning from Data
7.5 credits cr.
- Gå till denna sida på svenska webben
The course can be considered as a complement of the linear algebra courses you have studied at our department, but at a more advanced level.
We'll pay more attention on how to abstract relevant mathematics and structures from applications for example data and how to apply the theory you've studied. The course will start with some elementary elements in linear algebra e.g. SVD, principal components, matrix norms, generalized eigenvalues and interlacing eigenvalues. In particular we'll deal with these topics in a numerical sounding way. Later we'll turn to an interactive treatment of linear algebra and subjects from data.
- Basic computationally efficient algorithms for large matrices
- Principal Component Analysis
- Sparse and underdetermined systems and their relation to data compression
- Construction of neural networks and models for deep learning
- Fitting hyperparameters
- Selected topics on particular types of matrices
The course consists of one element.
Instruction consists of lectures, exercises, and computer projects.
The course is assessed through written examination and hand-in exercises.
A list of examiners can be found on
ScheduleThe schedule will be available no later than one month before the start of the course. We do not recommend print-outs as changes can occur. At the start of the course, your department will advise where you can find your schedule during the course.
Note that the course literature can be changed up to two months before the start of the course.
G. Strang: Linear Algebra and Learning from Data. Wellesley-Cambridge Press.