Invariance-based Multi-Clustering of Latent Space Embeddings for Equivariant Learning
Not on the map. Why?
Abstract (arXiv)
Variational Autoencoders (VAEs) have been shown to be remarkably effective in recovering model latent spaces for several computer vision tasks. However, currently trained VAEs, for a number of reasons, seem to fall short in learning invariant and equivariant clusters in latent space. Our work focuses on providing solutions to this problem and presents an approach to disentangle equivariance feature maps in a Lie group manifold by enforcing deep, group-invariant learning. Simultaneously implementing a novel separation of semantic and equivariant variables of the latent space representation, we formulate a modified Evidence Lower BOund (ELBO) by using a mixture model pdf like Gaussian mixtures for invariant cluster embeddings that allows superior unsupervised variational clustering. Our experiments show that this model effectively learns to disentangle the invariant and equivariant representations with significant improvements in the learning rate and an observably superior image recognition and canonical state reconstruction compared to the currently best deep learning models.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 23 of its 34 references that are in the map. First posted on arXiv 25 Jul 2021.
Builds on
23 of 34 references.
References · 23
- 2013Auto-Encoding Variational Bayescited by 4,053
- 2015Spatial Transformer Networkscited by 1,429
- 2016Group Equivariant Convolutional Networkscited by 1,165
- 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 860
- 2018Disentangling by Factorisingcited by 616
- 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 607
- 2018Understanding disentangling in β-VAEcited by 335
- 2016Harmonic Networks: Deep Translation and Rotation Equivariancecited by 295
- 2017VAE with a VampPriorcited by 248
- 2020Generalizing Convolutional Neural Networks for Equivariance to Lie Groups on Arbitrary Continuous Datacited by 248
- 2016Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoderscited by 220
- 2016Variational Deep Embedding: An Unsupervised and Generative Approach to Clusteringcited by 220
11 more
- 2018Isolating Sources of Disentanglement in Variational Autoencoderscited by 167
- 2015Deep Convolutional Inverse Graphics Networkcited by 121
- 2019AET vs. AED: Unsupervised Representation Learning by Auto-Encoding Transformations rather than Datacited by 77
- 2016Auxiliary Deep Generative Modelscited by 72
- 2018Learning Latent Subspaces in Variational Autoencoderscited by 51
- 2019Equivariant Transformer Networkscited by 35
- 2019Explicitly disentangling image content from translation and rotation with spatial-VAEcited by 28
- 2016A Survey of Inductive Biases for Factorial Representation-Learningcited by 23
- 2019AVT: Unsupervised Learning of Transformation Equivariant Representations by Autoencoding Variational Transformationscited by 15
- 2019Motion Equivariant Networks for Event Cameras with the Temporal Normalization Transformcited by 5
- 2019Affine Variational Autoencoders: An Efficient Approach for Improving Generalization and Robustness to Distribution Shiftcited by 1
Built on by
Citing papers per year: 2022: 1.
1 paper.