aiqbench analysis
This research proposes a novel framework for equivariant representation learning by decomposing latent spaces into invariant class factors and symmetry groups. By leveraging group theory to enforce lossless disentanglement between object identity and pose, the method offers improved interpretability over existing models on datasets with complex geometric symmetries.
Equivariant Representation Learning via Class-Pose Decomposition
arXiv:2207.03116v4 Announce Type: replace Abstract: We introduce a general method for learning representations that are equivariant to symmetries of data. Our central idea is to decompose the latent space into an invariant factor and the symmetry group itself. The components semantically correspond to intrinsic data classes and poses respectively. The learner is trained on a loss encouraging equivariance based on supervision from relative symmetry information. The approach is motivated by theoretical results from group theory and guarantees representations that are lossless, interpretable and disentangled. We provide an empirical investigation via experiments involving datasets with a variety of symmetries. Results show that our representations capture the geometry of data and outperform other equivariant representation learning frameworks.
Abstract:We introduce a general method for learning representations that are equivariant to symmetries of data. Our central idea is to decompose the latent space into an invariant factor and the symmetry group itself. The components semantically correspond to intrinsic data classes and poses respectively. The learner is trained on a loss encouraging equivariance based on supervision from relative symmetry information. The approach is motivated by theoretical results from group theory and guarantees representations that are lossless, interpretable and disentangled. We provide an empirical investigation via experiments involving datasets with a variety of symmetries. Results show that our representations capture the geometry of data and outperform other equivariant representation learning frameworks.
Machine Learning (cs.LG); Group Theory (math.GR)
arXiv:2207.03116 [cs.LG]
(or arXiv:2207.03116v4 [cs.LG] for this version)
Full text under the publisher's open licence · 124 words
Related coverage
Variational decomposition autoencoding improves disentanglement of latent representations
arXiv · Nov 2, 2026
Hamiltonian engineering via pulses: beyond group averaging
arXiv · Sep 17, 2026
MOSAIC: An Omni-Modal Benchmark for Long-Term Memory and Cross-Modal Reasoning in Conversational Agents
arXiv · Oct 22, 2026
Structured Learning on Mapper Representations
arXiv · Aug 25, 2026