Abstract:A rotationally invariant graph neural network (RotNet) is proposed for accurate and accelerated quantum mechanical calculations. Benchmark experiments demonstrate that RotNet outperforms prior state‐of‐the‐art methods and generalizes well to varying rotations. The high accuracy and efficiency suggest that RotNet has tremendous potential to significantly facilitate studies of protein dynamics simulation and materials engineering while maintaining physical plausibility. Deep learning has proven promising in biological and chemical applications, aiding in accurate predictions of properties such as atomic forces, energies, and material band gaps. Traditional methods with rotational invariance, one of the most crucial physical laws for predictions made by machine learning, have relied on Fourier transforms or specialized convolution filters, leading to complex model design and reduced accuracy and efficiency. However, models without rotational invariance exhibit poor generalization ability across datasets. Addressing this contradiction, this work proposes a rotationally invariant graph neural network, named RotNet, for accurate and accelerated quantum mechanical calculations that can overcome the generalization deficiency caused by rotations of molecules. RotNet ensures rotational invariance through an effective transformation and learns distance and angular information from atomic coordinates. Benchmark experiments on three datasets (protein fragments, electronic materials, and QM9) demonstrate that the proposed RotNet framework outperforms popular baselines and generalizes well to spatial data with varying rotations. The high accuracy, efficiency, and fast convergence of RotNet suggest that it has tremendous potential to significantly facilitate studies of protein dynamics simulation and materials engineering while maintaining physical plausibility.

The three-dimensional rotation neural network

The Storage Capacity Of The Complex Phasor Neural Network

A Rotational Motion Perception Neural Network Based on Asymmetric Spatiotemporal Visual Information Processing

Interpretable Rotation-Equivariant Quaternion Neural Networks for 3D Point Cloud Processing.

Energy Function for Learning Rotation Invariant in a Cascaded Neural Network Model

Rotation-Equivariant Quaternion Neural Networks for 3D Point Cloud Processing

The Hamilton Neural-Network Model

Bio-inspired visual neural network on spatio-temporal depth rotation perception

Cascaded model of neural networks and its learning algorithm for pattern recognition with rotation invariance

On the Continuity of Rotation Representations in Neural Networks

Vector-product Hopfield model

RotNet: A Rotationally Invariant Graph Neural Network for Quantum Mechanical Calculations

A Feed-Forward Neural Network for Increasing the Hopfield-Network Storage Capacity

Dynamical analysis and DSP implementation of 3D Hopfield neural network under dual memristive electromagnetic radiation

Probabilistic Rotation Modeling Based on Directional Mixture Density Networks

Nonergodic delocalized paramagnetic states in quantum neural networks

A Triple-Memristor Hopfield Neural Network With Space Multistructure Attractors and Space Initial-Offset Behaviors

Neural network models for the quaternion singular value decompositions

Complex Rotation Quantum Dynamic Neural Networks (CRQDNN) using Complex Quantum Neuron (CQN): Applications to time series prediction.

Rotation-Invariant Point Cloud Representation for 3-D Model Recognition

On Representation of 3D Rotation in the Context of Deep Learning