Scale-Equivariant Steerable NetworksIvan SosnovikMicha SzmajaArnold SmeuldersInternational Conference on Learning Representations
First, we introduce the general theory for building scale-equivariant convolutional networks with steerable filters. We develop scale-convolution and generalize other common blocks to be scale-equivariant. We demonstrate the computational efficiency and numerical stability of the proposed method. We compare...
atom-centered message-passing interatomic potentials in particular, each atom’s hidden latent space is a feature vector consisting solely of invariant scalars25. More recently, however, a class of models known as equivariant neural networks33,34,35,36have been ...
Scale-Equivariant Steerable Networks. In Proceedings of the 8th International Conference on Learning Representations (ICLR), Addis Ababa, Ethiopia, 26–30 April 2020. [Google Scholar] Xu, Y.; Xiao, T.; Zhang, J.; Yang, K.; Zhang, Z. Scale-Invariant Convolutional Neural Networks. arXiv ...
Group equivariant convolutional networks. In Proceedings of the 33rd International Conference on Machine Learning (ICML 2016), New York City, NY, USA, 19–24 June 2016; pp. 2990–2999. [Google Scholar] Andrearczyk, V.; Fageot, J.; Oreiller, V.; Montet, X.; Depeursinge, A. Local ...