Equivariant Representations for Non-Free Group Actions

We introduce a method for learning representations that are equivariant with respect to general group actions over data. Differently from existing equivariant representation learners, our method is suitable for actions that are not free i.e., that stabilize data via nontrivial symmetries. Our method is grounded in the orbit-stabilizer theorem from group theory, which guarantees that an ideal learner infers an isomorphic representation. Finally, we provide an empirical investigation on image datasets with rotational symmetries and show that taking stabilizers into account improves the quality of the representations.

[1]  D. Kragic,et al.  Equivariant Representation Learning via Class-Pose Decomposition , 2022, AISTATS.

[2]  Danilo Jimenez Rezende,et al.  Symmetry-Based Representations for Artificial and Biological General Intelligence , 2022, Frontiers in Computational Neuroscience.

[3]  Yoshua Bengio,et al.  Properties from Mechanisms: An Equivariance Perspective on Identifiable Representation Learning , 2021, ICLR.

[4]  J. Portegies,et al.  Quantifying and Learning Linear Symmetry-Based Disentanglement , 2020, ICML.

[5]  Geoffrey E. Hinton,et al.  A Simple Framework for Contrastive Learning of Visual Representations , 2020, ICML.

[6]  David Filliat,et al.  Symmetry-Based Disentangled Representation Learning requires Interaction with Environments , 2019, NeurIPS.

[7]  David Pfau,et al.  Towards a Definition of Disentangled Representations , 2018, ArXiv.

[8]  Frank Hutter,et al.  Decoupled Weight Decay Regularization , 2017, ICLR.

[9]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  Jianxiong Xiao,et al.  3D ShapeNets: A deep representation for volumetric shapes , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  Max Welling,et al.  Learning the Irreducible Representations of Commutative Lie Groups , 2014, ICML.

[12]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  William R. Clements,et al.  Learning Disentangled Representations and Group Structure of Dynamical Environments , 2020, NeurIPS.

[14]  George W. Polites,et al.  An introduction to the theory of groups , 1968 .

[15]  J. Rotman.  An Introduction to the Theory of Groups , 1965 .