DGPE PhD course: Machine Learning for Health

11-13 June 2019. Lecturers: Jon Kolstad and Ziad Obermeyer

2019.03.22 | Susanne Christensen

Date Tue 11 Jun Thu 13 Jun
Time 09:00    17:00
Location University of Southern Denmark

Course information

This course will prepare students to:

  • Describe intuitively how and why a few basic machine learning algorithms work
  • Understand what is new about data science and how it differs from traditional estimation (regression). For example, distinguish prediction problems (‘can I predict y with x’) from estimation problems (‘does x cause y ’)
  • Identify major methodological pitfalls encountered in answering these kinds of questions
  • Develop a research question around a prediction policy problem in health, and make a research plan that avoids key methodological problems

Course evaluation will be based on in-class participation, as well as a short research proposal that applies machine learning methods to a problem in health. Proposals will be evaluated on the basis of how well students incorporate methodological insights from the course, their ability to recognize and work around pitfalls, and the originality of the proposal.

Course Outline 



Students who would like to participate in the course should register through the link found on University of Southern Denmark's website: https://www.sdu.dk/da/om_sdu/institutter_centre/ivoe_virksomhedsledelse_og_oekonomi/forskning/satsningsomraader/nhe_homepage/nhenews/4

Registration will open on May 1.

Course Fees

The course is free of charge for PhD students in Social Sciences from Danish PhD programs. PhD students in Social Sciences from other universities need pay a tuition of 300 euros. The tuition covers refreshments and lunch on all days as well as a workshop dinner on June 11, 2019. Participants are expected to cover their own transportation and accommodation costs (if needed).

ECTS Points

Upon completing all course activities, participants will be awarded 3 ECTS credits and a course certificate.


Course Coordinator: N. Meltem Daysal, University of Southern Denmark 
Administrative Support: Helle Møller Jensen, University of Southern Denmark

DGPE Courses