DeepSphere: towards an equivariant graph-based spherical CNN

Abstract

Spherical data is found in many applications. By modeling the discretized sphere as a graph, we can accommodate non-uniformly distributed, partial, and changing samplings. Moreover, graph convolutions are computationally more efficient than spherical convolutions. As equivariance is desired to exploit rotational symmetries, we discuss how to approach rotation equivariance using the graph neural network introduced in Defferrard et al. (2016). Experiments show good performance on rotation-invariant learning problems. Code and examples are available at https://github.com/SwissDataScienceCenter/DeepSphere

published at the ICLR 2019 Workshop on Representation Learning on Graphs and Manifolds. arXiv admin note: text overlap with arXiv:1810.12186

References in corpus (10)

• HEALPix -- a Framework for High Resolution Discretization, and Fast Analysis of Data Distributed on the Sphere
• The Emerging Field of Signal Processing on Graphs: Extending High-Dimensional Data Analysis to Networks and Other Irregular Domains
• Group Equivariant Convolutional Networks
• On the Generalization of Equivariance and Convolution in Neural Networks to the Action of Compact Groups
• Stationary signal processing on graphs
• DeepSphere: Efficient spherical Convolutional Neural Network with HEALPix sampling for cosmological applications
• Gauss--Legendre Sky Pixelization (GLESP) for CMB maps
• Learning Spherical Convolution for Fast Features from 360° Imagery
• Spherical CNNs on Unstructured Grids
• Cosmological constraints from the convergence 1-point probability distribution

Cited by in corpus (6)

• A neural simulation-based inference approach for characterizing the Galactic Center $\gamma$-ray excess
• Constant Curvature Graph Convolutional Networks
• DeepSphere: a graph-based spherical CNN
• Group Equivariant Conditional Neural Processes
• Curvature Graph Neural Network
• Hermitian Symmetric Spaces for Graph Embeddings