Capsule Network Projectors are Equivariant and Invariant Learners
- URL: http://arxiv.org/abs/2405.14386v1
- Date: Thu, 23 May 2024 10:04:23 GMT
- Title: Capsule Network Projectors are Equivariant and Invariant Learners
- Authors: Miles Everett, Aiden Durrant, Mingjun Zhong, Georgios Leontidis,
- Abstract summary: In this work, we propose an invariant-equivariant self-supervised architecture that employs Capsule Networks (CapsNets)
We demonstrate that the use of CapsNets in equivariant self-supervised architectures achieves improved downstream performance.
This approach, which we name CapsIE (Capsule Invariant Equivariant Network), achieves state-of-the-art performance.
- Score: 4.909818180516128
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Learning invariant representations has been the longstanding approach to self-supervised learning. However, recently progress has been made in preserving equivariant properties in representations, yet do so with highly prescribed architectures. In this work, we propose an invariant-equivariant self-supervised architecture that employs Capsule Networks (CapsNets) which have been shown to capture equivariance with respect to novel viewpoints. We demonstrate that the use of CapsNets in equivariant self-supervised architectures achieves improved downstream performance on equivariant tasks with higher efficiency and fewer network parameters. To accommodate the architectural changes of CapsNets, we introduce a new objective function based on entropy minimisation. This approach, which we name CapsIE (Capsule Invariant Equivariant Network), achieves state-of-the-art performance across all invariant and equivariant downstream tasks on the 3DIEBench dataset, while outperforming supervised baselines. Our results demonstrate the ability of CapsNets to learn complex and generalised representations for large-scale, multi-task datasets compared to previous CapsNet benchmarks. Code is available at https://github.com/AberdeenML/CapsIE.
Related papers
- RobCaps: Evaluating the Robustness of Capsule Networks against Affine
Transformations and Adversarial Attacks [11.302789770501303]
Capsule Networks (CapsNets) are able to hierarchically preserve the pose relationships between multiple objects for image classification tasks.
In this paper, we evaluate different factors affecting the robustness of CapsNets, compared to traditional Conal Neural Networks (CNNs)
arXiv Detail & Related papers (2023-04-08T09:58:35Z) - 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) - Self-supervised learning of Split Invariant Equivariant representations [0.0]
We introduce 3DIEBench, consisting of renderings from 3D models over 55 classes and more than 2.5 million images where we have full control on the transformations applied to the objects.
We introduce a predictor architecture based on hypernetworks to learn equivariant representations with no possible collapse to invariance.
We introduce SIE (Split Invariant-Equivariant) which combines the hypernetwork-based predictor with representations split in two parts, one invariant, the other equivariant, to learn richer representations.
arXiv Detail & Related papers (2023-02-14T07:53:18Z) - Capsules as viewpoint learners for human pose estimation [4.246061945756033]
We show how most neural networks are not able to generalize well when the camera is subject to significant viewpoint changes.
We propose a novel end-to-end viewpoint-equivariant capsule autoencoder that employs a fast Variational Bayes routing and matrix capsules.
We achieve state-of-the-art results for multiple tasks and datasets while retaining other desirable properties.
arXiv Detail & Related papers (2023-02-13T09:01:46Z) - Equivariant Transduction through Invariant Alignment [71.45263447328374]
We introduce a novel group-equivariant architecture that incorporates a group-in hard alignment mechanism.
We find that our network's structure allows it to develop stronger equivariant properties than existing group-equivariant approaches.
We additionally find that it outperforms previous group-equivariant networks empirically on the SCAN task.
arXiv Detail & Related papers (2022-09-22T11:19:45Z) - SVNet: Where SO(3) Equivariance Meets Binarization on Point Cloud
Representation [65.4396959244269]
The paper tackles the challenge by designing a general framework to construct 3D learning architectures.
The proposed approach can be applied to general backbones like PointNet and DGCNN.
Experiments on ModelNet40, ShapeNet, and the real-world dataset ScanObjectNN, demonstrated that the method achieves a great trade-off between efficiency, rotation, and accuracy.
arXiv Detail & Related papers (2022-09-13T12:12:19Z) - Semantic-aware Modular Capsule Routing for Visual Question Answering [55.03883681191765]
We propose a Semantic-aware modUlar caPsulE framework, termed as SUPER, to better capture the instance-specific vision-semantic characteristics.
We comparatively justify the effectiveness and generalization ability of our proposed SUPER scheme over five benchmark datasets.
arXiv Detail & Related papers (2022-07-21T10:48:37Z) - Improving the Sample-Complexity of Deep Classification Networks with
Invariant Integration [77.99182201815763]
Leveraging prior knowledge on intraclass variance due to transformations is a powerful method to improve the sample complexity of deep neural networks.
We propose a novel monomial selection algorithm based on pruning methods to allow an application to more complex problems.
We demonstrate the improved sample complexity on the Rotated-MNIST, SVHN and CIFAR-10 datasets.
arXiv Detail & Related papers (2022-02-08T16:16:11Z) - Efficient-CapsNet: Capsule Network with Self-Attention Routing [0.0]
Deep convolutional neural networks make extensive use of data augmentation techniques and layers with a high number of feature maps to embed object transformations.
capsule networks are a promising solution to extend current convolutional networks and endow artificial visual perception with a process to encode more efficiently all feature affine transformations.
In this paper, we investigate the efficiency of capsule networks and, pushing their capacity to the limits with an extreme architecture with barely 160K parameters, we prove that the proposed architecture is still able to achieve state-of-the-art results.
arXiv Detail & Related papers (2021-01-29T09:56:44Z) - Interpretable Graph Capsule Networks for Object Recognition [17.62514568986647]
We propose interpretable Graph Capsule Networks (GraCapsNets), where we replace the routing part with a multi-head attention-based Graph Pooling approach.
GraCapsNets achieve better classification performance with fewer parameters and better adversarial robustness, when compared to CapsNets.
arXiv Detail & Related papers (2020-12-03T03:18:00Z) - Invariant Integration in Deep Convolutional Feature Space [77.99182201815763]
We show how to incorporate prior knowledge to a deep neural network architecture in a principled manner.
We report state-of-the-art performance on the Rotated-MNIST dataset.
arXiv Detail & Related papers (2020-04-20T09:45:43Z)
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.