Postdoc position ”REMAX: Rigorous Evaluation Methods for AI Explainability” at Computer Science, Aarhus University Full-time NEW

Postdoc Position “REMAX: Rigorous Evaluation Methods for AI Explainability” at Computer Science, Aarhus University

The Department of Computer Science at Aarhus University invites applications for a 24-month Postdoctoral Research Fellow in Explainable AI interested in interdisciplinary research between computer science and philosophy.

The REMAX project on Rigorous Evaluation Methods for AI Explainability, funded by a Villum Synergy grant, aims to create methods for explainable AI and their rigorous evaluation. It is a collaboration between the Department of Computer Science and the Centre for Science Studies at Aarhus University, Denmark.

The position is available from 1 December 2026, or as soon as possible thereafter.

Explaining complex AI models is a key challenge for ethically responsible AI. Explainable AI (XAI) research aims to provide relevant information to assist developers and users in analyzing AI models.

However, rigorous methods for evaluating XAI algorithms are currently lacking. This is in large part due to a gap between how explainability is evaluated within computer science and philosophy. In computer science, explainability is mainly evaluated through data-driven metrics. While they provide some insights for developers, it is unclear whether existing metrics track any ethically relevant properties. Conversely, in philosophy, explainability is evaluated using normative criteria. While based on principled ethical analyses of why explainability matters, they are currently too abstract and provide little practical guidance for AI development or governance. To bridge this gap, our goal is to: (a) pioneer new normative criteria for AI explainability, drawing on theories of evidence and explanation in philosophy of science; (b) develop novel metrics and algorithms for AI explainability; and (c) create a tight feedback loop where (a) and (b) can iteratively refine each other.

As a postdoc, you will be based in the Data-Intensive Systems research group, under the guidance of Prof. Ira Assent, and part of the ADA (Algorithms, Data, and Artificial Intelligence) in the Department of Computer Science. You will draw on your expertise in models and algorithms for AI and XAI, to engage with our colleagues in Science Studies, headed by Assoc. Prof. Rune Nyrup, where you will have a secondary office space.

Aarhus University is the second-oldest and second-largest university in Denmark. Aarhus is the second-largest city in Denmark and is situated by the sea in East Jutland. Aarhus is a “young” city due to its proportionally large population of young inhabitants. Denmark offers an attractive work-life balance and work environment.

Applicants must hold or expect to complete a relevant PhD degree with documented expertise in devising and analyzing AI models, preferably including XAI methods, with a track record including publications in renowned venues of the Machine Learning and Data Mining field.

Your application should include a CV, list of publications, letters of recommendation, as well as a brief letter of motivation. The letter should include a description of your background as relevant to the REMAX project, a motivation for interdisciplinary collaboration, and a sketch of potential research ideas if any (in total max. 1000 words).

Applications must be received no later than 30 September 2026.

The application must be submitted via Aarhus University's recruitment system, which can be accessed under the job advertisement on Aarhus University's website.