English

Roto-translated Local Coordinate Frames For Interacting Dynamical Systems

Machine Learning 2024-03-21 v3 Machine Learning

Abstract

Modelling interactions is critical in learning complex dynamical systems, namely systems of interacting objects with highly non-linear and time-dependent behaviour. A large class of such systems can be formalized as geometric graphs\textit{geometric graphs}, i.e.\textit{i.e.}, graphs with nodes positioned in the Euclidean space given an arbitrarily\textit{arbitrarily} chosen global coordinate system, for instance vehicles in a traffic scene. Notwithstanding the arbitrary global coordinate system, the governing dynamics of the respective dynamical systems are invariant to rotations and translations, also known as Galilean invariance\textit{Galilean invariance}. As ignoring these invariances leads to worse generalization, in this work we propose local coordinate frames per node-object to induce roto-translation invariance to the geometric graph of the interacting dynamical system. Further, the local coordinate frames allow for a natural definition of anisotropic filtering in graph neural networks. Experiments in traffic scenes, 3D motion capture, and colliding particles demonstrate that the proposed approach comfortably outperforms the recent state-of-the-art.

Keywords

Cite

@article{arxiv.2110.14961,
  title  = {Roto-translated Local Coordinate Frames For Interacting Dynamical Systems},
  author = {Miltiadis Kofinas and Naveen Shankar Nagaraja and Efstratios Gavves},
  journal= {arXiv preprint arXiv:2110.14961},
  year   = {2024}
}

Comments

In NeurIPS 2021. Source code: https://github.com/mkofinas/locs

R2 v1 2026-06-24T07:15:28.689Z