球形卷积神经网络的扩展

摘要

球形卷积神经网络将卷积神经网络推广到球面上的函数，通过使用球形卷积作为主要线性操作。计算球形卷积最准确和高效的方法是在频谱域中（通过卷积定理），但仍然比通常的平面卷积更昂贵。因此，迄今为止，球形卷积神经网络的应用仅限于可以用较低模型容量解决的小问题。在这项工作中，我们展示了如何将球形卷积神经网络扩展到规模更大的问题。为实现这一目标，我们进行了关键改进，包括常见模型组件的新变体、实现核心操作以利用硬件加速器特性，以及利用我们模型特性的特定应用输入表示。实验表明，我们的更大规模球形卷积神经网络在QM9分子基准的多个目标上达到了最先进水平，该基准以前主要由等变图神经网络主导，并在多个天气预测任务上取得了竞争性表现。我们的代码可在https://github.com/google-research/spherical-cnn 上找到。

English

Spherical CNNs generalize CNNs to functions on the sphere, by using spherical convolutions as the main linear operation. The most accurate and efficient way to compute spherical convolutions is in the spectral domain (via the convolution theorem), which is still costlier than the usual planar convolutions. For this reason, applications of spherical CNNs have so far been limited to small problems that can be approached with low model capacity. In this work, we show how spherical CNNs can be scaled for much larger problems. To achieve this, we make critical improvements including novel variants of common model components, an implementation of core operations to exploit hardware accelerator characteristics, and application-specific input representations that exploit the properties of our model. Experiments show our larger spherical CNNs reach state-of-the-art on several targets of the QM9 molecular benchmark, which was previously dominated by equivariant graph neural networks, and achieve competitive performance on multiple weather forecasting tasks. Our code is available at https://github.com/google-research/spherical-cnn.

球形卷积神经网络的扩展

Scaling Spherical CNNs

摘要

Support