Using and Abusing Equivariance
- URL: http://arxiv.org/abs/2308.11316v1
- Date: Tue, 22 Aug 2023 09:49:26 GMT
- Title: Using and Abusing Equivariance
- Authors: Tom Edixhoven, Attila Lengyel, Jan van Gemert
- Abstract summary: We show how Group Equivariant Convolutional Neural Networks use subsampling to learn to break equivariance to their symmetries.
We show that a change in the input dimension of a network as small as a single pixel can be enough for commonly used architectures to become approximately equivariant, rather than exactly.
- Score: 10.70891251559827
- License: http://creativecommons.org/licenses/by-sa/4.0/
- Abstract: In this paper we show how Group Equivariant Convolutional Neural Networks use
subsampling to learn to break equivariance to their symmetries. We focus on 2D
rotations and reflections and investigate the impact of broken equivariance on
network performance. We show that a change in the input dimension of a network
as small as a single pixel can be enough for commonly used architectures to
become approximately equivariant, rather than exactly. We investigate the
impact of networks not being exactly equivariant and find that approximately
equivariant networks generalise significantly worse to unseen symmetries
compared to their exactly equivariant counterparts. However, when the
symmetries in the training data are not identical to the symmetries of the
network, we find that approximately equivariant networks are able to relax
their own equivariant constraints, causing them to match or outperform exactly
equivariant networks on common benchmark datasets.
Related papers
- A Probabilistic Approach to Learning the Degree of Equivariance in Steerable CNNs [5.141137421503899]
Steerable convolutional neural networks (SCNNs) enhance task performance by modelling geometric symmetries.
Yet, unknown or varying symmetries can lead to overconstrained weights and decreased performance.
This paper introduces a probabilistic method to learn the degree of equivariance in SCNNs.
arXiv Detail & Related papers (2024-06-06T10:45:19Z) - Learning Layer-wise Equivariances Automatically using Gradients [66.81218780702125]
Convolutions encode equivariance symmetries into neural networks leading to better generalisation performance.
symmetries provide fixed hard constraints on the functions a network can represent, need to be specified in advance, and can not be adapted.
Our goal is to allow flexible symmetry constraints that can automatically be learned from data using gradients.
arXiv Detail & Related papers (2023-10-09T20:22:43Z) - Investigating how ReLU-networks encode symmetries [13.935148870831396]
We investigate whether equivariance of a network implies that all layers are equivariant.
We conjecture that CNNs trained to be equivariant will exhibit layerwise equivariance.
We show that it is typically easier to merge a network with a group-transformed version of itself than merging two different networks.
arXiv Detail & Related papers (2023-05-26T15:23:20Z) - Optimization Dynamics of Equivariant and Augmented Neural Networks [2.7918308693131135]
We investigate the optimization of neural networks on symmetric data.
We compare the strategy of constraining the architecture to be equivariant to that of using data augmentation.
Our analysis reveals that even in the latter situation, stationary points may be unstable for augmented training although they are stable for the manifestly equivariant models.
arXiv Detail & Related papers (2023-03-23T17:26:12Z) - Self-Supervised Learning for Group Equivariant Neural Networks [75.62232699377877]
Group equivariant neural networks are the models whose structure is restricted to commute with the transformations on the input.
We propose two concepts for self-supervised tasks: equivariant pretext labels and invariant contrastive loss.
Experiments on standard image recognition benchmarks demonstrate that the equivariant neural networks exploit the proposed self-supervised tasks.
arXiv Detail & Related papers (2023-03-08T08:11:26Z) - Deep Neural Networks with Efficient Guaranteed Invariances [77.99182201815763]
We address the problem of improving the performance and in particular the sample complexity of deep neural networks.
Group-equivariant convolutions are a popular approach to obtain equivariant representations.
We propose a multi-stream architecture, where each stream is invariant to a different transformation.
arXiv Detail & Related papers (2023-03-02T20:44:45Z) - The Lie Derivative for Measuring Learned Equivariance [84.29366874540217]
We study the equivariance properties of hundreds of pretrained models, spanning CNNs, transformers, and Mixer architectures.
We find that many violations of equivariance can be linked to spatial aliasing in ubiquitous network layers, such as pointwise non-linearities.
For example, transformers can be more equivariant than convolutional neural networks after training.
arXiv Detail & Related papers (2022-10-06T15:20:55Z) - Learning Symmetric Embeddings for Equivariant World Models [9.781637768189158]
We propose learning symmetric embedding networks (SENs) that encode an input space (e.g. images)
This network can be trained end-to-end with an equivariant task network to learn an explicitly symmetric representation.
Our experiments demonstrate that SENs facilitate the application of equivariant networks to data with complex symmetry representations.
arXiv Detail & Related papers (2022-04-24T22:31:52Z) - Equivariance versus Augmentation for Spherical Images [0.7388859384645262]
We analyze the role of rotational equivariance in convolutional neural networks (CNNs) applied to spherical images.
We compare the performance of the group equivariant networks known as S2CNNs and standard non-equivariant CNNs trained with an increasing amount of data augmentation.
arXiv Detail & Related papers (2022-02-08T16:49:30Z) - Group Equivariant Subsampling [60.53371517247382]
Subsampling is used in convolutional neural networks (CNNs) in the form of pooling or strided convolutions.
We first introduce translation equivariant subsampling/upsampling layers that can be used to construct exact translation equivariant CNNs.
We then generalise these layers beyond translations to general groups, thus proposing group equivariant subsampling/upsampling.
arXiv Detail & Related papers (2021-06-10T16:14:00Z) - Learning Invariances in Neural Networks [51.20867785006147]
We show how to parameterize a distribution over augmentations and optimize the training loss simultaneously with respect to the network parameters and augmentation parameters.
We can recover the correct set and extent of invariances on image classification, regression, segmentation, and molecular property prediction from a large space of augmentations.
arXiv Detail & Related papers (2020-10-22T17:18:48Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.