Structuring Representation Geometry with Rotationally Equivariant Contrastive Learning

Open in new window