Symmetry-Based Disentangled Representation Learning requires Interaction with Environments

2019/03/30 by Caselles-Dupré, Hugo, Garcia-Ortiz, Michael, Filliat, David · 2 citations
#FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.1904.00243

Abstract

Finding a generally accepted formal definition of a disentangled representation in the context of an agent behaving in an environment is an important challenge towards the construction of data-efficient autonomous agents. Higgins et al. recently proposed Symmetry-Based Disentangled Representation Learning, a definition based on a characterization of symmetries in the environment using group theory. We build on their work and make observations, theoretical and empirical, that lead us to argue that Symmetry-Based Disentangled Representation Learning cannot only be based on static observations: agents should interact with the environment to discover its symmetries. Our experiments can be reproduced in Colab and the code is available on GitHub.

Cited by

• Developmental Symmetry-Loss: A Free-Energy Perspective on Brain-Inspired Invariance Learning
• What Drives Compositional Generalization in Visual Generative Models?

Related

• Towards a Definition of Disentangled Representations
• Learning Group Structure and Disentangled Representations of Dynamical Environments
• Algebras of actions in an agent's representations of the world
• Quantifying and Learning Linear Symmetry-Based Disentanglement
• On the Symmetries of Deep Learning Models and their Internal Representations