Explainers
Short pieces on the ideas behind our research, most with an animation. They started as posts on LinkedIn and have been edited for the web.
Why a CNN beats a vision transformer on small data
Every symmetry you build into an architecture is data you don’t have to collect.
What masked reconstruction teaches a model about shape
Hide part of a shape, ask a model to rebuild it, and symmetry becomes a shortcut worth learning.
Noether’s theorem, in one spring
Every symmetry comes with something that never changes. A spring shows how.
The 6-or-9 problem
Why world models trained to compress what they see can learn the wrong thing about the world.
Gaussian, Dirichlet and Beta processes
Three Bayesian models for when you have little data, and what they’re used for today.
No trade-off between energy and force accuracy
Neural surrogates for chemistry are tuned as if energy and force accuracy compete. On two molecules, they didn’t.
Solomonoff, and why simpler is usually better
One of the first theories of machine learning was a theory of compression.