<ul><li>Equivariant neural networks aim to improve performance by incorporating symmetries through group actions.</li><li>A new zero-parameter approach is proposed to impose approximate equivariance for a finite group in the latent representation.</li><li>Experiments show that the network learns a group representation on the latent space and prefers to learn the regular representation.</li><li>The proposed approach is benchmarked on three datasets and shows similar or better performance compared to existing equivariant methods with fewer parameters.</li></ul>

Parameter-free approximate equivariance for tasks with finite group symmetry

Discover more