"One should never try to prove anything that is not almost obvious".
Alexander Grothendieck
I am a doctoral student at the Royal Institute of Technology (KTH) in Stockholm, Sweden. I pursue research in geometric methods for Machine Learning, Statistics and Data Science.
Research
Spaces of data naturally carry intrinsic geometry. The latter is a rich structure encoding (potentially the entirety of) semantics. In the words of Kepler,
"Where there is matter data, there is geometry".
Below you can find a selection of my academic works. Click on an image to get to the corresponding arXiv entry. For a complete list please visit my Google Scholar profile.
An Efficient and Continuous Voronoi Density Estimator
Equivariant Representation Learning via Class-Pose Decomposition
AISTATS 2023Learning Geometric Representations of Objects via Interaction
ECML-PKDD 2023Voronoi Density Estimator for High-Dimensional Data: Computation, Compactification and Convergence
UAI 2022Harmonics of Learning: Universal Fourier Features Emerge in Invariant Networks
PreprintBack to the Manifold: Recovering from Out-of-Distribution States
IROS 2022Hearts and Towers in Stable ∞-Categories
Journal of Homotopy 14Resume
Please click here to download my academic resume.
Blog
Under construction.