Machine learning
The course addresses components of machine learning such as data, hypothesis space and loss functions and it also introduces algorithms for machine learning: gradient descent, greedy search and linear solvers.
Schedule:
–
Teaching time:
Daytime
Location:
Online
Topic:
Information and communications technology
Form of learning:
Online
Provider:
Aalto University, FITech
Level:
Intermediate
Credits:
5 By Aalto University (ECTS)
Fee:
Free of charge
Target group and prerequisites
Matrix algebra, probability theory, basic programming skills.
Course description
Course contents
- Components of machine learning: Data, hypothesis space and loss functions
- Algorithms for machine learning: gradient descent, greedy search, linear solvers
Learning outcomes
After completing the course, the students
- can formalise applications as ML problems and solve them using basic ML methods.
- understand the concept of generalisation and how to analyse it using simple probabilistic models.
- are familiar with linear models for regression and classification.
- know how basic ML methods are obtained as combinations of particular choices for data representation (features), hypothesis space (model) and loss function.
- are familiar with the idea of hard and soft clustering methods.
- understand the basic idea of feature learning methods.
Teaching methods
The course follows a schedule and includes lectures, self study, assignments, and a project work.
The lectures might be organised on campus but will in any case be available online.
Workload
5 credits, approx. 130 hours of work divided into:
- lectures + self-study (30 hours)
- assignments (6 * 10 = 60 hours)
- project work (around 40 hours)
Application deadline: 2.1.2022.
