Group equivariant convolutional neural networks (G-CNNs) have been successfully applied in geometric deep learning. Typically, G-CNNs have the advantage over CNNs that they do not waste network capacity on training symmetries that should have been hard-coded in the network. The recently introduced framework of PDE-based G-CNNs (PDE-G-CNNs) generalises G-CNNs. PDE-G-CNNs have the core advantages that they simultaneously 1) reduce network complexity, 2) increase classification performance, and 3) provide geometric interpretability. Their implementations primarily consist of linear and morphological convolutions with kernels. In this paper we show that the previously suggested approximative morphological kernels do not always accurately approximate the exact kernels accurately. More specifically, depending on the spatial anisotropy of the Riemannian metric, we argue that one must resort to sub-Riemannian approximations. We solve this problem by providing a new approximative kernel that works regardless of the anisotropy. We provide new theorems with better error estimates of the approximative kernels, and prove that they all carry the same reflectional symmetries as the exact ones. We test the effectiveness of multiple approximative kernels within the PDE-G-CNN framework on two datasets, and observe an improvement with the new approximative kernels. We report that the PDE-G-CNNs again allow for a considerable reduction of network complexity while having comparable or better performance than G-CNNs and CNNs on the two datasets. Moreover, PDE-G-CNNs have the advantage of better geometric interpretability over G-CNNs, as the morphological kernels are related to association fields from neurogeometry.
翻译:等变群卷积神经网络(G-CNNs)已成功应用于几何深度学习领域。通常情况下,G-CNNs相较于CNNs具有优势,因为它们不会在训练网络对称性时浪费网络容量。最近引进的基于偏微分方程的G-CNNs (PDE-G-CNNs)扩展了G-CNNs,PDE-G-CNNs的核心优势是它们可以同时实现以下三个目的: 1) 减少网络复杂度,2) 提高分类性能, 3) 提供几何可解释性。它们的实现主要包括线性和形态学卷积。在本文中,我们表明先前提出的近似形态学核不总是准确的逼近,尤其是取决于Riemannian测度的空间各向异性。我们认为必须采用亚里曼尼逼近方法,并提供了适用于任何各向异性情况下的新近似核。我们提供具有更好误差估计的新定理,证明它们与精确的核具有相同的反射对称性。我们在两个数据集上测试了PDE-G-CNNs框架内多个逼近核的有效性,并观察到使用新逼近核的性能提高。我们报告PDE-G-CNNs再次允许在比G-CNNs和CNNs更少复杂度的情况下具有与这两个方法相当或更优的性能。此外,PDE-G-CNNs具有比G-CNNs更好的几何可解释性,因为形态学核与神经几何学中的关联场有关。