Equivariance Through ParameterSharing
[edit]
Proceedings of the 34th International Conference on Machine Learning, PMLR 70:28922901, 2017.
Abstract
We propose to study equivariance in deep neural networks through parameter symmetries. In particular, given a group G that acts discretely on the input and output of a standard neural network layer, we show that its equivariance is linked to the symmetry group of network parameters. We then propose two parametersharing scheme to induce the desirable symmetry on the parameters of the neural network. Under some conditions on the action of G, our procedure for tying the parameters achieves Gequivariance and guarantees sensitivity to all other permutation groups outside of G.
Related Material


