Kaisa_2012_3_photo by Veikko Somerpuro

Enrol

Timetable

Here is the course’s teaching schedule. Check the description for possible other schedules.

DateTimeLocation
Mon 28.10.2019
10:15 - 12:00
Mon 28.10.2019
12:15 - 14:00
Thu 31.10.2019
10:15 - 12:00
Mon 4.11.2019
10:15 - 12:00
Mon 4.11.2019
12:15 - 14:00
Thu 7.11.2019
10:15 - 12:00
Mon 11.11.2019
10:15 - 12:00
Mon 11.11.2019
12:15 - 14:00
Thu 14.11.2019
10:15 - 12:00
Mon 18.11.2019
10:15 - 12:00
Mon 18.11.2019
12:15 - 14:00
Thu 21.11.2019
10:15 - 12:00
Mon 25.11.2019
10:15 - 12:00
Mon 25.11.2019
12:15 - 14:00
Thu 28.11.2019
10:15 - 12:00
Mon 2.12.2019
10:15 - 12:00
Mon 2.12.2019
12:15 - 14:00
Thu 5.12.2019
10:15 - 12:00
Mon 9.12.2019
10:15 - 12:00
Mon 9.12.2019
12:15 - 14:00
Thu 12.12.2019
10:15 - 12:00

Other teaching

Conduct of the course

All course materials are in Moodle.

Grading:

* Pre-quizzes for lecture material (20%)
* 3 projects (70%): Privacy 1 (20%), Privacy 2 (20%), Fairness (30%)
* Lecture attendance (10%)

Pre-quizzes

* Small Moodle quizzes with multiple-choice questions about material to be read for each lecture
* Deadline: at 10:00 before a lecture starting at 10:15
* Passing each quiz is a requirement for completing the course
* Missed or failed quizzes must be compensated by writing a short (~1-2 page) report on the same material

More details about the projects will be added later

Description

Master's Programme in Data Science is responsible for the course.

The course is available to students from other degree programmes.

Prerequisites in terms of knowledge

Prerequisites for students in the Data Science programme, in terms of courses

Introduction to Machine Learning

Introduction to Data Science

Prerequisites for other students in terms of courses

Introduction to Machine Learning

Introduction to Data Science

Recommended preceding courses

Advanced Course in Machine Learning

LDA-C5009 Philosophy of Artificial Intelligence
SOSM-326 Datafication - critical perspectives

After the course, the student:

  • Knows different aspects of trustworthy machine learning as a basis for ethical machine learning and AI
  • Understands the principles and most important variants of differential privacy as a basis for privacy-preserving machine learning
  • Can explain the main challenges in defining and measuring fairness in machine learning
  • Can identify main types of techniques for algorithm sanitization, understand their limitations and strengths, can implement and apply them
  • Can apply basic techniques for privacy-preserving and fair machine learning

Second year of Data Science MSc studies.

Autumn term, Period II

The course will cover methods for trustworthy and ethical machine learning and AI, focusing on the technical perspective of methods that allow addressing current ethical issues for example in privacy and fairness. We will focus more on major design principles of the solutions rather than detailed mathematical theory. Nonetheless, basic working knowledge of machine learning methods is a strict requirement as a prerequisite.

Literature is based on research articles and other online material and will be provided during the course.

During the lectures we will cover material from research articles and the students are expected to have read the articles before the lecture so that they can participate in class discussions.

Exercises in the course will mainly focus on experimenting with privacy-preserving and fair machine learning in practice and applying them to concrete data science problems. There will be weekly exercise sessions for discussions around the problems and Q&A sessions.

Grading scale 0-5

Grading will be based on a combination of course exam, mandatory course exercises and online mini-exams, as well as additional exercises as given during the course.

The course will consist of lectures, on-line mini-exams, programming exercises, and possibly other forms of teaching.

Activity during the course, including possibly mandatory attendance, will be required to pass the course.

Antti Honkela and Indrė Žliobaitė