PhD position in Hybrid Intelligence: Explaining data-driven decisions with legal, ethical or social impact to end-users (1.0 FTE)
Updated: 04 Aug 2020
Why build AI systems that replace people if we can build AI systems which collaborate with people? Hybrid Intelligence is the combination of human and machine intelligence, expanding human intellect instead of replacing it. It is our goal to design Hybrid Intelligent systems, an approach to Artificial Intelligence that puts humans at the centre, changing the course of the ongoing AI revolution. The project will be recruiting 15 PhD or postdoc positions in total, at six Universities across the Netherlands. More information on the project can be found here.At Utrecht University we are looking for a PhD candidate who is interested in explaining outcomes of data-driven machine-learning applications that support decision making procedures to end-users of these applications, such as lawyers, business people or ordinary citizens. The techniques should apply in contexts where a human decision maker is informed by data-driven algorithms and where the decisions have ethical, legal or societal implications. They should generate explanations for outputs for specific inputs. The explanations generated should be such that the reasons for the output can be understood and critically examined on their quality.The project will specifically focus on explaining ‘black-box’ applications in that it will focus on model-agnostic methods, assuming only access to the training data and the possibility to evaluate a model’s output, given the input data. This will make the explanation methods independent of a model’s internal structure. This is important since in many real-life applications the learned models will not be interpretable or accessible, for instance, when the model is learned by deep learning or when the application is proprietary.The project will explore the use of existing and novel techniques of formal and computational argumentation. It will in particular study techniques for case based argumentation, which model how decision makers draw analogies to past cases (precedents) and discuss their relevant similarities and differences. A case based approach is natural since the training data of (supervised) machine-learning applications can be seen as decided past cases, i.e. as precedents. An argumentation based approach is important, since the focus is not just on understanding outputs but also on critically examining their quality.As a PhD candidate you will be part of the Department of Information and Computing Sciences (Intelligent Systems group) and you will work in close partnership with the Department of Artificial Intelligence of the Faculty of Science and Engineering of the University of Groningen.Daily supervision will be performed by Prof Henry Prakken (UU), Prof Davide Grossi (RUG) and Prof Bart Verheij (RUG).
You are an enthusiastic, creative and highly motivated candidate who has:
- a Master's degree in Artificial Intelligence, Computer Science, Mathematics, Data Science or a related field;
- a strong interest in interdisciplinary AI research that combines the AI subfields of machine learning, data science, knowledge representation & reasoning and human-computer interaction;
- the ability to integrate various research methods and tools, such as formal methods, designing and implementing algorithms and experimental evaluation;
- a commitment to develop computational tools and techniques in order to help people make better decisions;
- excellent English communication skills (written and spoken);
- a wish to contribute to the department’s teaching programmes (15% at most);
- a talent for planning and organizing your work, and for meeting deadlines.
- a full-time position for 4 years;
- access to a network of top AI researchers through the Hybrid Intelligence Centre;
- participation in the education programme of the Graduate School of Natural Sciences, as well as in the Hybrid Intelligence project Graduate School;
- a full-time gross salary starts at €2,395 and increases to €3,061 per month in the fourth year (scale P of the Collective Labour Agreement Dutch Universities);
- benefits including 8% holiday bonus and 8.3% end-of-year bonus;
- a pension scheme, a partially paid parental leave and flexible employment conditions (based on the Collective Labour Agreement Dutch Universities (cao)).
In addition to the employment conditions laid down in the cao for Dutch Universities, Utrecht University has a number of its own arrangements. For example, there are agreements on professional development, leave arrangements and sports. We also give you the opportunity to expand your terms of employment yourself via the Employment Conditions Selection Model. This way we like to encourage you to continue to grow.
More information about working at the Faculty of Science can be found here.
38 - 40 hours per week