PhD student in causality, machine learning and computational social science

Linkoping

Reference number IDA-2024-00311

We have the power of over 40,000 students and co-workers. Students who provide hope for the future. Co-workers who contribute to Linköping University meeting the challenges of the day. Our fundamental values rest on credibility, trust and security. By having the courage to think freely and innovate, our actions together, large and small, contribute to a better world. We look forward to receiving your application!

We are looking for a PhD student in causality, machine learning and computational social science

Your work assignments

A central challenge of the social sciences is that society is a complex system due to its intricate interdependencies. For instance, individuals' lives are shaped by exposures and interactions within and across social domains such as workplaces and neighborhoods. While this complexity is acknowledged in theory, much of the empirical work tends to simplify it. This risks leading to an overly simplified description of the social world. Recent developments in machine learning (ML) paired with rich population data, however, offer a path forward for the computational social sciences. The aim of this project is to make use of and further develop the most recent ML methods for analyzing population registers.

Secifically, this project aims to estimate the causal effects of exposures such as neighborhood disadvantage on life opportunities and outcomes, a.k.a. contextual effects. To this end, we aim to make use of transformer neural networks, which are designed to efficiently model complex dependencies. While these models have revolutionized the analysis of textual and image, its application in the computational social sciences is still in its infancy.

The key challenge in estimating contextual effects is distinguishing them from confounding effects stemming from individuals non-randomly selecting into contexts. For this, recent ML approaches have sought to proxy unobserved confounders based on observed selection behaviors and social positions. Transformers thus have considerable potential in improving existing proxy-based approaches. To this end, two considerations are of utmost importance. (1) There are several conceivable ways that the embeddings could be used, and so a first task is to explore which is the most appropriate. (2) As we have shown in previous works, using proxies come with their own challenges, and must be derived carefully to avoid introducing bias. We will identify conditions and criteria under which unbiased estimation is possible and develop sensitivity analyses attuned to these criteria.

As a PhD student, you devote most of your time to doctoral studies and the research described above. Your work may also include teaching or other departmental duties, up to a maximum of 20% of full-time. The work assignments also include actively contributing to the collaborative environment within which the project will be carried out (read more under “Your workplace” below).

Your qualifications

You have graduated at master’s level in machine learning, statistics, computer science, computational social science or a related area that is considered relevant for the research topic of the project, or have completed courses with a minimum of 240 credits, at least 60 of which must be in advanced courses in the subject areas mentioned above. Alternatively, you have gained essentially corresponding knowledge in another way.

A successful candidate should have excellent study results and a strong background in mathematics. The applicant should be skilled at implementing new models and algorithms in a suitable software environment, with documented experience. The applicant should furthermore have a strong drive towards performing fundamental research; the ability and interest to work collaboratively; and strong communication skills. The applicant should be able to communicate freely in oral and written English.

Your workplace

Linköping University is one of the leading AI institutions in Sweden. We have strong links to prominent national research initiatives, such as WASP, ELLIIT and SweCSS and you will have access to state-of-the-art computing infrastructure for machine learning, e.g. through BerzeLiUs.

The position is formally based at the Division of Statistics and Machine Learning (STIMA) within the Department of Computer and Information Science. At STIMA we conduct research and education in both statistics and machine learning, at the undergraduate, advanced and PhD levels. STIMA is part of the Department of Computer and Information Science (IDA) – a dynamic, international, and collaborative environment with a strong focus on research, which spans the range from purely theoretical to applied.

STIMA is a part of the Department of Computer and Information Science (IDA). Read more about the department here: https://liu.se/en/organisation/liu/ida 

This project will be a close collaboration between STIMA (main supervisor: Jose M. Peña, senior associate professor in machine learning), and the Institute for Analytical Sociology (co-supervisor: Adel Daoud, senior associate professor in social science), and the Swedish Excellence Centre for Computational Social Science.

The employment

When taking up the post, you will be admitted to the program for doctoral studies. More information about the doctoral studies at each faculty is available at Doctoral studies at Linköping University.

The employment has a duration of four years’ full-time equivalent. You will initially be employed for a period of one year. The employment will subsequently be renewed for periods of maximum duration two years, depending on your progress through the study plan. The employment may be extended up to a maximum of five years, based on the amount of teaching and departmental duties you have carried out. Further extensions can be granted in special circumstances.

Starting date by agreement.

Salary and employment benefits

The salary of PhD students is determined according to a locally negotiated salary progression.

More information about employment benefits at Linköping University is available here.

Union representatives

Information about union representatives, see Help for applicants.

Application procedure

Apply for the position by clicking the “Apply” button below. Your application must reach Linköping University no later than December 10, 2024. Applications and documents received after the date above will not be considered.

N.B. When applying for the position we want you to provide a personal letter (first field in the application form). This letter should contain a paragraph where you briefly explain/list the qualifications that you believe are particularly relevant for the research project described above. This paragraph should start with the words “Suitability for research topic:”.

We welcome applicants with different backgrounds, experiences and perspectives - diversity enriches our work and helps us grow. Preserving everybody's equal value, rights and opportunities is a natural part of who we are. Read more about our work with: Equal opportunities.




We look forward to receiving your application!





Linköping university has framework agreements and wishes to decline direct contacts from staffing- and recruitment companies as well as vendors of job advertisements.

Contact persons

Jose M. Peña

Senior Associate Professor

jose.m.pena@liu.se

Sofie Bondesson

HR Administartor

sofie.bondesson@liu.se


URL to this page
https://web103.reachmee.com/ext/I011/853/main?site=7&validator=d7a66c13be778ef950c393a904293789&lang=UK&rmpage=job&rmjob=25648&rmlang=UK