Postdoc in developing deep learning for satellite images

This is a call for one postdoctoral position in the Data Science and AI division at the Department of Computer Science and Engineering, Chalmers University of Technology. The positions are situated within the research project Observatory of Poverty, situated in the AI and Global Development Lab (you will find more information about the Lab at www.global-lab.ai), funded by the Swedish Research Council (SRC). The Lab furthers the use of AI for Social Good for the sustainable development goals.

Project description
the main challenge that the postdoc is expected to contribute to.

1.To train learning algorithms to estimate poverty from satellite images, of African communities over time and space, quarterly, from 1984 to 2020.
2.To contribute to creating a statistical package—ObservatoryOfPoverty—that enables us, and other scholars, to produce poverty estimates for policy evaluation.

Another goal of the Observatory of Poverty is to use our poverty estimates to identify to what extent African communities are trapped in poverty and explain how competing aid and development interventions alter these communities’ prospects to free themselves from deprivation. This goal is about causal inference. Thus, a secondary goal of the postdoc position is to facilitate the methodological development for causal inference for the project.

In other words, we are searching for a candidate that is interested in either poverty estimation from satellite images or the causal inference part of the project, or both. Your interest should be clearly stated in your application and linked to your track-record.

The candidate for the postdoc position will join the AI and Global Development Lab and is expected to produce research that contributes to the listed objectives or related spin-off objectives (e.g., based on the candidate’s research interest). Such spin-offs are welcomed, especially those that provide a new research angle to the listed objectives.

As the project is highly interdisciplinary, we welcome applicants from a variety of disciplinary backgrounds, such as computational social science, statistics, computer science, mathematics, information technology, computer or electrical engineering, signal processing, physics or related fields, and experience in machine learning.

We publish in top interdisciplinary generalist journals, discipline-specific journals, and conferences. We will adapt our publication strategy depending on the candidate’s background and interest.

Regardless of background, to contribute to this research, the candidate should have experience in the following items:


• Excellent communication skills in written and spoken English.
• Excellent coding skills in Python, R, or equivalent language
• Experience in using deep learning.
• Experience in image processing, preferably satellite images

The candidate should have a proven track record in applying deep-learning methods with publications in peer-reviewed journals or conferences.

More informtion about the project and mentorship is found here

Work environment
The AI and Global Development Lab tend to meet weekly and remotely, with collaborators based in Sweden, the United States, India, Chile, United Kingdom.

The project constitutes a collaboration mainly among the Department of Engineering and Computer Science, Chalmers University of Technology, the Departments of Political Science, University of Gothenburg, Institute for Analytical Sociology, Linköping University (campus Norrköping) and the Department of Statistics, Harvard University. This means that occasional travels within Sweden and between Sweden and the United States are expected.

The candidate is encouraged to check out the YouTube playlist (provided here) and some of the recent interviews are the following:
• Pursuing excellence. An interview with Professor Gary King, Harvard University. (Link to the show)
• How to combine academia and entrepreneurship. Continuing the conversation with Prof. King at Harvard. (Link to the show)
• Following your curiosity. Interview with Xiao-Li Meng, Professor of Statistics, Harvard University. (Link to the show)
• Establishing your research program. An Interview with Prof. Stephen Raudenbush University of Chicago. (Link to the show)
• Choosing your academic path. Interview with Professor Christopher Winship, Harvard University. (Link to the show)
• Building excellent research environments. Interview with Professor Peter Hedström, Linköping University. (Link to the show)
• Finding one's path as a statistician or data scientist. An Interview with Prof. Jennifer Hill, New York University. (Link to the show)

Major responsibilities
The main responsibility of the candidate will be to pursue research in the area of the position. The candidate will have the chance to work with the group members and collaborators at Chalmers and abroad, including the co-supervision of PhD students. The candidate is encouraged to supervise/teach up to 20% of their time.

Contract terms
Full-time temporary employment. The employment is limited (temporary) for 24 months with the possibility of extention with 12 months (contingent on funding and research performance).

Qualifications
Applicants must have a doctorate, PhD or equivalent degree, in a relevant field.  The degree should generally not be older than three years, but exception can be made.

To qualify, candidates must have experience in one or more of (1) deep learning, (2) causal inference (3) image data (preferably earth observations), (4) modeling geo-temporal data, as evidence by publications in appropriate venues. 

The position requires sound verbal and written communication skills in English. Swedish is not a requirement at Chalmers.

More information and apply here

Application deadline: February 20 , 2023

For questions, please contact:
Adel Daoud, CSE DSAI
E-mail: daoud@chalmers.se

*** Chalmers declines to consider all offers of further announcement publishing or other types of support for the recruiting process in connection with this position. ***