I am an Applied Scientist at Amazon Web Services where I work on Responsible AI. Before, I did a (short) postdoc doing probabilistic machine learning at the Technical University of Denmark, Section for Cognitive Systems. This is also where I got my PhD with Søren Hauberg as my advisor. During my PhD I spent some time with Mark van der Wilk 's group at Imperial College London and James Hensman's team at Amazon Cambridge. This is a personal website: opinions are my own and not representative of AWS.
-
Geographical Erasure in Language Generation
Our EMNLP 2023 Findings paper finds that LLMs are not always geographically inclusive.
-
Papers
Three preprints have turned into conference papers this year.
-
PhD Defence
I successfully defended my PhD.
-
AI Ethics Reading List
I put together a list of (popular) science books and blog posts.
-
Gradient Article
We wrote an article on algoritmic fairness for the Gradient.
-
Paper
A preprint for "Last Layer Marginal Likelihood for Invariance Learning" is out on arxiv.
-
Thoughts on Algorithmic Fairness
Can we get machines to make just decisions, and should we try?
-
Paper
The "Probabilistic Spatial Transformers for Bayesian Data Augmentation" preprint is out on arxiv.
-
Talk at AI Expo
I gave a talk at DTU's Skylab - "Bias in - Bias out? Building Fair Models from Imbalanced Data."
-
Summer School Organization
I hosted a summer school on fairness and machine learning.