A team led by Imperial College Professor and Twitter Chief Scientist Michael Bronstein publishes a paper that aims to geometrically unify the typical architectures of CNNs, GNNs, LSTMs, Transformers, etc. from the perspective of symmetry and invariance to build an “Erlangen Programme” for deep neural networks.

Here is a quick read: Twitter Tech Lead Michael Bronstein & Team Leverage the Erlangen Programme to Establish the Geometric Foundations of Deep Learning.

The paper Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges is on arXiv.

Source link