Exploiting Learned Symmetries in Group Equivariant Convolutions
Group Equivariant Convolutions (GConvs) enable convolutional neural networks to be equivariant to various transformation groups, but at an additional parameter and compute cost. We investigate the filter parameters learned by GConvs and find certain conditions under which they become highly redundant. We show that GConvs can be efficiently decomposed into depthwise separable convolutions while preserving equivariance properties and demonstrate improved performance and data efficiency on two datasets. All code is publicly available at github.com/Attila94/SepGrouPy.
Code (1)
Similar Papers 제목 키워드 기반
Group Equivariant Convolutional Networks
We introduce Group equivariant Convolutional Neural Networks (G-CNNs), a natural generalization of convolutional neural networks that reduces sample complexity by exploiting symmetries. G-CNNs use G-convolutions, a new t…
Breast Tumour ClassificationColorectal Gland Segmentation:Multi-tissue Nucleus SegmentationRotated MNISTEfficient Continuous Group Convolutions for Local SE(3) Equivariance in 3D Point Clouds
Extending the translation equivariance property of convolutional neural networks to larger symmetry groups has been shown to reduce sample complexity and enable more discriminative feature learning. Further, exploiting a…
Semantic SegmentationTranslationSymmetry From Scratch: Group Equivariance as a Supervised Learning Task
In machine learning datasets with symmetries, the paradigm for backward compatibility with symmetry-breaking has been to relax equivariant architectural constraints, engineering extra weights to differentiate symmetries …
Inductive BiasCurrent Symmetry Group Equivariant Convolution Frameworks for Representation Learning
Euclidean deep learning is often inadequate for addressing real-world signals where the representation space is irregular and curved with complex topologies. Interpreting the geometric properties of such feature spaces h…
Deep LearningRepresentation LearningSO(3)-Equivariant Neural Networks for Learning Vector Fields on Spheres
Analyzing vector fields on the sphere, such as wind speed and direction on Earth, is a difficult task. Models should respect both the rotational symmetries of the sphere and the inherent symmetries of the vector fields. …