Approximately equivariant graph networks

N Huang, R Levie, S Villar - Advances in Neural …, 2024 - proceedings.neurips.cc
Graph neural networks (GNNs) are commonly described as being permutation equivariant
with respect to node relabeling in the graph. This symmetry of GNNs is often compared to …

Invariant slot attention: Object discovery with slot-centric reference frames

O Biza, S Van Steenkiste, MSM Sajjadi… - ar**
X Ma, X Fu, T Wang, L Zhuo, Q Zou - Bioinformatics, 2024 - academic.oup.com
Motivation Accurate prediction of acute dermal toxicity (ADT) is essential for the safe and
effective development of contact drugs. Currently, graph neural networks, a form of deep …

Newton–cotes graph neural networks: On the time evolution of dynamic systems

L Guo, W Wang, Z Chen, N Zhang… - Advances in …, 2024 - proceedings.neurips.cc
Abstract Reasoning system dynamics is one of the most important analytical approaches for
many scientific studies. With the initial state of a system as input, the recent graph neural …

Equivariant graph neural operator for modeling 3d dynamics

M Xu, J Han, A Lou, J Kossaifi, A Ramanathan… - arxiv preprint arxiv …, 2024 - arxiv.org
Modeling the complex three-dimensional (3D) dynamics of relational systems is an
important problem in the natural sciences, with applications ranging from molecular …

Relaxing continuous constraints of equivariant graph neural networks for broad physical dynamics learning

Z Zheng, Y Liu, J Li, J Yao, Y Rong - Proceedings of the 30th ACM …, 2024 - dl.acm.org
Incorporating Euclidean symmetries (eg rotation equivariance) as inductive biases into
graph neural networks has improved their generalization ability and data efficiency in …

Geometric trajectory diffusion models

J Han, M Xu, A Lou, H Ye, S Ermon - arxiv preprint arxiv:2410.13027, 2024 - arxiv.org
Generative models have shown great promise in generating 3D geometric systems, which is
a fundamental problem in many natural science domains such as molecule and protein …

Dusego: Dual second-order equivariant graph ordinary differential equation

Y Wang, N Yin, M **ao, X Yi, S Liu, S Liang - arxiv preprint arxiv …, 2024 - arxiv.org
Graph Neural Networks (GNNs) with equivariant properties have achieved significant
success in modeling complex dynamic systems and molecular properties. However, their …

Pose-Transformed Equivariant Network for 3D Point Trajectory Prediction

R Yu, J Sun - Proceedings of the IEEE/CVF Conference on …, 2024 - openaccess.thecvf.com
Predicting 3D point trajectory is a fundamental learning task which commonly should be
equivariant under Euclidean transformation eg SE (3). The existing equivariant models are …