I’m a Master’s student in Artificial Intelligence at the University of Amsterdam. My goal is to have a positive impact on the world by making sure that AI will be aligned with humanity’s interests.
I’m currently working on my Master’s thesis with the Center for Human-Compatible AI on distance measures between reward functions. Before that, I have done work on the interpretability of reward models, equivariant deep learning, and on graph-based segmentation.
MSc in Artificial Intelligence, since 2020
University of Amsterdam
BSc in Physics, 2020
We prove impossibility results showing that Karger’s contraction algorithm cannot be extended to $s$-$t$-mincuts or normalized cuts. However, we show how extensions of Karger’s algorithm can still be useful for seeded segmentation.
We present a method for simplifying a learned reward model before visualizing it and show that this can make the reward more interpretable.