Postdoc in Bias and Fairness

Wednesday 10 Jun 20

Apply for this job

Apply no later than 1 August 2020
Apply for the job at DTU Compute by completing the following form.

Apply online

As a postdoc in bias and fairness for medicine, you will be part of the section for Image Analysis and Computer Graphics at DTU Compute. We are a lively research environment dedicated to machine learning, statistics and algorithms to solve methodological and applied problems within imaging and computer graphics. You will also be affiliated with the Neurobiology Research Unit (NRU) at Rigshospitalet (www.nru.dk), a cross-disciplinary research center focusing on brain neurobiology and advanced data analytic approaches of neuroimaging data.

The research project Bias and Fairness in Medicine aims to develop new concepts of fairness for medicine in a collaboration that involves both machine learning, medicine, and ethics.

While state-of-the-art fair algorithms largely focus on mathematically constraining predictive algorithms to be fair in a legal sense, these constraints are often bound to lead to decreased predictive power, which may be problematic in medicine: Are we willing to reduce our ability to diagnose depression in females just because we are less able to diagnose it in males? In collaboration with ethics, you will formulate new concepts of fairness and derive mathematical models for fair AI algorithms in medicine. 

Responsibilities and tasks 
The goal of the project is to collaboration with researchers in ethics for AI to formulate new concepts of fairness specialized for medicine; derive mathematical models for how predictive algorithms will be fair in this new sense; and to document the effect of dataset bias on fair versus un-corrected algorithms via a registry study of diagnostic bias in depression. 

In particular:
 

  • You will contribute to a registry study on depression, documenting bias in diagnosis on the dataset, and documenting how the dataset bias propagates to predictive algorithms trained on the dataset.
  • You will develop mathematical models of fairness within medicine in collaboration with researchers within ethics for AI
  • You will develop geometric transfer learning models based on e.g. optimal transport, and assessing their potential for improving fairness for medical AI
  • Your research could include popular scientific dissemination to the general public and popular media.
This is an interdisciplinary research project involving three research groups: The group of Aasa Feragen (DTU Compute) which specializes in machine learning for biomedical imaging and related problems; the group of Melanie Ganz (NRU) which specializes in statistical methods for neuroimaging, and the group of Sune Hannibal Holm (University of Copenhagen), which specializes in ethics for AI. While your base will be at DTU Compute, you will be co-affiliated with NRU, where the registry study will be located. You will also collaborate actively with the ethics group, in particular for deriving new, ethically motivated, concepts of fairness.

Qualifications
Candidates should have a PhD degree in computer science, statistics, mathematics or a similar degree. Additionally, you should have:  
  • An unstoppable drive and excitement for developing responsible machine learning algorithms
  • A documented solid background in either machine learning, statistics, or geometry
  • Programming experience including at least one of Python, Matlab, or R
The position is part of a collaborative, interdisciplinary project. You thrive in collaboration with others, and you are excited to interact with researchers from different research fields and turn their problems into machine learning.

You are ambitious and wish to make a difference in how AI is used in healthcare. You seek to publish your research at top venues, and you are eager to communicate your science broadly to maximize its impact.

We offer
DTU is a leading technical university globally recognized for the excellence of its research, education, innovation and scientific advice. We offer a rewarding and challenging job in an international environment. We strive for academic excellence in an environment characterized by collegial respect and academic freedom tempered by responsibility. 

Salary and terms of employment
The appointment will be based on the collective agreement with the Danish Confederation of Professional Associations. The allowance will be agreed upon with the relevant union.

The period of employment is 2 years, starting October 1 2020 or as soon as possible thereafter. Your workplace will be split between the DTU Lyngby Campus and the Neurobiology Research Unit at Rigshospitalet in Copenhagen.

You can read more about
career paths at DTU here.

Further information 
Further information may be obtained from Aasa Feragen, tel.: +45 2622 0498, email afhar@dtu.dk

You can read more about the project at
fairmed.compute.dtu.dk, and you can read more about DTU Compute at www.compute.dtu.dk/english and aout NRU at www.nru.dk

Application procedure 
Please submit your online application no later than 1 August 2020 (23:59 local time)
Applications must be submitted as one PDF file containing all materials to be given consideration. To apply, please open the link "Apply online", fill out the online application form, and attach all your materials in English in one PDF file. The file must include: 
  • Application (cover letter)
  • CV
  • Diploma (MSc/PhD)
  • List of publications 
Applications and enclosures received after the deadline will not be considered.

All interested candidates irrespective of age, gender, disability, race, religion or ethnic background are encouraged to apply.

DTU Compute conducts research and provides teaching in the fields of mathematics, modeling and computer science. The expanding mass of information and the increasingly complex use of advanced technology in society demand development of advanced computer based mathematical models and calculations. The unique skills of the department are in high demand in IT innovation and production.

Technology for people
DTU develops technology for people. With our international elite research and study programmes, we are helping to create a better world and to solve the global challenges formulated in the UN’s 17 Sustainable Development Goals. Hans Christian Ørsted founded DTU in 1829 with a clear vision to develop and create value using science and engineering to benefit society. That vision lives on today. DTU has 11,500 students and 6,000 employees. We work in an international atmosphere and have an inclusive, evolving, and informal working environment. Our main campus is in Kgs. Lyngby north of Copenhagen and we have campuses in Roskilde and Ballerup and in Sisimiut in Greenland.