PhD Position in Causality-Aware Explanations for Probabilistic Graphical Models

Effective collaboration between humans and AI in high-stake domains with inherent uncertainty requires clear insights into the effects of actions and strategies for improvement.​ Meaningful causal explanations can help the human to accept the AI’s advice. The aim of this study is to Integrate causality into probabilistic model explanations to improve human understanding and trust in AI systems.​

Your job

Probabilistic models are central to modern AI systems, particularly in domains where uncertainty is the norm. In high-stake domains, decisions supported through AI systems can have a significant impact and users therefore need to be able to contest and understand how the decisions are made. Uncertainty underlying predictions and advice is hard to explain, but causality is a concept closer to human intuition.

In this project, you will study uncertainty and causality in probabilistic graphical models with the goal of designing causality-aware explanations. Explanations should also be adaptable for different users who may have different views on what counts as a meaningful explanation. Your research will be part of the Hybrid Intelligence Centre, which aims to design systems that profit from both human and machine intelligence. This large nation-wide consortium provides you with numerous opportunities to collaborate with a diverse group of PhD candidates and more senior researchers on various challenges and use cases, both within and outside our department.You will be employed by Utrecht University and have the status of an employee rather than a student status. You will become a member of the Intelligent Systems group of the Department of Information and Computing Sciences (ICS) and a regular visitor of the Leiden Institute of Advanced Computer Science (LIACS) at Leiden University. Daily supervision will be performed by dr. Silja Renooij (UU-ICS) and dr. Anna V. Kononova (LIACS).

Your main tasks and opportunities:

• conduct original research related to the above mentioned topics, using a combination of techniques ranging from mathematics, algorithm design and implementation, to computer experiments and design research;
• publish, present and share your findings through scientific articles and at international journals and conferences;
• actively participate in the Hybrid Intelligence consortium and take part in its training programme, as well as the educational programme of the Graduate School of Natural Sciences;
• contribute to the teaching activities of the ICS department (10-15% of your time), gaining valuable academic experience;
• write and defend your PhD thesis within the duration of your contract.

Your qualities

We are looking for a creative and responsible candidate who is highly motivated to develop into an independent researcher. You bring the skills and critical mindset needed to make a valuable contribution to education and research in computer science and hybrid intelligence. More specifically, you fit to this role if you:

• have an MSc degree in Computer Science or Artificial Intelligence with a strong mathematical background, or an MSc degree in (Applied) Mathematics or Statistics, or an MSc degree in a related field with a comparable background;
• are a skilled programmer (Python and/or C++), demonstrated via relevant courses or projects;
• have at least basic knowledge and understanding of probabilistic graphical models and/or causal inference;
• are proficient in English (spoken and written);
• have demonstrable skills in planning your work effectively, prioritizing tasks, and meeting deadlines.

We especially encourage you to apply if you:

• can demonstrate an interest in human-centred AI, human-centred design, or related subjects, e.g. via relevant courses or projects;
• are interested in combining applied and theoretical research in an interdisciplinary context, with societal impact;
• are willing to actively collaborate within and contribute to the Hybrid Intelligence (HI) consortium, its challenges and specifically the Diabetes use case;

We aim to create hybrid intelligence for everyone, see also our Diversity Statement. To do this, we need an inclusive and diverse team of researchers. We especially encourage people from underrepresented groups to apply for this job.

Our offer

• A position for 18 months, with an extension to a total of four years upon successful assessment in the first 18 months;
• a gross monthly salary between €2,901 and €3,707 in the case of full-time employment (salary scale P under the Collective Labour Agreement for Dutch Universities (CAO NU));
• 8% holiday pay and 8.3% year-end bonus;
• a pension scheme, partially paid parental leave and flexible terms of employment based on the CAO NU.