The research group Responsible Artificial Intelligence (RAI) at the Department of Computing Science at Umeå University is seeking outstanding candidates for a post-doctoral position in Computer Science with focus on formal verification of algorithmic decision-making systems.
Automated decision‐making (ADM) systems are increasingly extending and replacing human decisions in a large number of situations areas, with significant impacts on people’s lives and society. The responsible design and use of these systems requires methods and tools to support accountability and the verification of how decisions are aligned with human rights and democratic principles.
This project extends our current work on Glass Box methods, focussing on the formal specification and verification of ethical principles. More precisely, the postdoc is expected to develop formal models and theories that, given a set of features (representing ethical values and/or societal requirements) can determine how to mathematically represent those features to verify the alignment of an ADM. The project is expected to deliver a sound theoretical framework to guide the verification of ADM systems and the choice of formal accountability provisions.