PhD Position on Explainable AI for Public Policy Design

The lack of transparency hinders the use of AI techniques in real decision-support contexts since results obtained from black-box optimization methods are generally not trusted. We would not expect decision makers to leverage policy recommendations when those recommendations cannot be fully explained.

This PhD project will focus on enhancing the interpretability and explainability of bio-inspired heuristic optimization methods. In particular, this project will formalize the patterns and pathways of operators when searching for efficient tradeoffs. Additionally, it will set foundations to assess interpretability, as currently there is ambiguity about what exactly constitutes interpretability in heuristic methods. Further, we will advance knowledge about the trade-offs between interpretability and predictive accuracy. 

In short, this project will build on the increasing body of literature on explainable AI, but contribute a new research line on interpretable and explainable heuristic methods with the aim of improving accountability of AI-based decision support.

This is a PhD position in the Hippo (Hyper-heuristics for interpretable public policy analysis) lab. The core goal of the Hippo lab is to advance multi-purpose decision support using collaborative and heuristic AI methods. You will work as part of an interdisciplinary team between the Interactive Intelligence section in the Intelligent Systems (INSY) department at the Faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS) and the Policy Analysis section in the Multi-Actor systems (MAS) department at the Faculty of Technology Policy and Management (TPM), tackling the following major scientific challenges:

The Hippo lab is a Delft Artificial Intelligence Lab. Artificial Intelligence, Data and Digitalisation are becoming increasingly important when looking for answers to major scientific and societal challenges. In a DAI-lab, experts in ‘the fundamentals of AI technology’ along with experts in ‘AI challenges’ run a shared lab. As a PhD, you will work with at least two academic members of staff and three other PhD candidates. In total TU Delft will establish 24 DAI-Labs, where 48 Tenure Trackers and 96 PhD candidates will have the opportunity to push the boundaries of science using AI. Each team is driven by research questions which arise from scientific and societal challenges, and contribute to the development and execution of domain specific education. Data and Digitalisation education related activities. All team members have many opportunities for self-development. You will be a member of the thriving DAI-Lab community that fosters cross-fertilization between talents with different expertise and disciplines.

For description of other projects in Hippo lab, visit: https://ii.tudelft.nl/~pradeep/hippo.html