e3nn: Euclidean Neural Networks
arXiv:2207.09453
Abstract
We present e3nn, a generalized framework for creating E(3) equivariant trainable functions, also known as Euclidean neural networks. e3nn naturally operates on geometry and geometric tensors that describe systems in 3D and transform predictably under a change of coordinate system. The core of e3nn are equivariant operations such as the TensorProduct class or the spherical harmonics functions that can be composed to create more complex modules such as convolutions and attention mechanisms. These core operations of e3nn can be used to efficiently articulate Tensor Field Networks, 3D Steerable CNNs, Clebsch-Gordan Networks, SE(3) Transformers and other E(3) equivariant networks.
arXiv commentdraft
Cited by in corpus (19)
- A Comprehensive Survey on Deep Graph Representation Learning
- Force-Field-Enhanced Neural Network Interactions: from Local Equivariant Embedding to Atom-in-Molecule properties and long-range effects
- Molecular dynamics simulation with finite electric fields using Perturbed Neural Network Potentials
- Physics-Driven Learning for Inverse Problems in Quantum Chromodynamics
- An equivariant graph neural network for the elasticity tensors of all seven crystal systems
- Artificial Intelligence in Materials Science and Engineering: Current Landscape, Key Challenges, and Future Trajectorie
- Lorentz group equivariant autoencoders
- Is novelty predictable?
- Machine learning and invariant theory
- Structure discovery in Atomic Force Microscopy imaging of ice
- Equivariant Neural Networks for Spin Dynamics Simulations of Itinerant Magnets
- Beam induced heating in electron microscopy modeled with machine learning interatomic potentials
- Generative Model for Constructing Reaction Path from Initial to Final States
- EOSnet: Embedded Overlap Structures for Graph Neural Networks in Predicting Material Properties
- A Cartesian Encoding Graph Neural Network for Crystal Structures Property Prediction: Application to Thermal Ellipsoid Estimation
- Fine-Tuning Unifies Foundational Machine-learned Interatomic Potential Architectures at ab initio Accuracy
- Energy-GNoME: A Living Database of Selected Materials for Energy Applications
- Leveraging SO(3)-steerable convolutions for pose-robust semantic segmentation in 3D medical data
- Probing the Temporal Response of Liquid Water to a THz Pump Pulse Using Machine Learning-Accelerated Non-Equilibrium Molecular Dynamics