Efficient Generalized Spherical CNNs

Abstract

Many problems across computer vision and the natural sciences require the analysis of spherical data, for which representations may be learned efficiently by encoding equivariance to rotational symmetries. We present a generalized spherical CNN framework that encompasses various existing approaches and allows them to be leveraged alongside each other. The only existing non-linear spherical CNN layer that is strictly equivariant has complexity $\mathcal{O}(C^2{L}^5)$, where $C$ is a measure of representational capacity and $L$ the spherical harmonic bandlimit. Such a high computational cost often prohibits the use of strictly equivariant spherical CNNs. We develop two new strictly equivariant layers with reduced complexity $\mathcal{O}(C{L}^4)$ and $\mathcal{O}(C{L}^3\log L)$, making larger, more expressive models computationally feasible. Moreover, we adopt efficient sampling theory to achieve further computational savings. We show that these developments allow the construction of more expressive hybrid models that achieve state-of-the-art accuracy and parameter efficiency on spherical benchmark problems.

arXiv comment20 pages, 4 figures, accepted by ICLR, code at https://www.kagenova.com/products/fourpiAI/

References in corpus (15)

• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds
• On the Generalization of Equivariance and Convolution in Neural Networks to the Action of Compact Groups
• DeepSphere: Efficient spherical Convolutional Neural Network with HEALPix sampling for cosmological applications
• Gauge Equivariant Convolutional Networks and the Icosahedral CNN
• A novel sampling theorem on the sphere
• An icosahedron-based method for pixelizing the celestial sphere
• Spherical CNNs
• Fast Convolution on the Sphere
• Fast directional continuous spherical wavelet transform algorithms
• Localisation of directional scale-discretised wavelets on the sphere
• Directional spin wavelets on the sphere
• A novel sampling theorem on the rotation group
• Theoretical Aspects of Group Equivariant Neural Networks
• Spin-Weighted Spherical CNNs