AI & ML Paradigm Shift

Enables agents to autonomously discover the group structure of their environments to learn disentangled representations without human priors.

arXiv · March 13, 2026 · 2603.11790

Dang-Nhu Barthélémy, Annabi Louis, Argentieri Sylvain

Why it matters

Traditional symmetry-based disentanglement requires researchers to pre-define environment groups (e.g., Euclidean rotations). This method allows an embodied agent to learn these structures unsupervised from its own action data, enabling models to build more robust world representations in unknown domains.

From the abstract

Symmetry-based disentangled representation learning leverages the group structure of environment transformations to uncover the latent factors of variation. Prior approaches to symmetry-based disentanglement have required strong prior knowledge of the symmetry group's structure, or restrictive assumptions about the subgroup properties. In this work, we remove these constraints by proposing a method whereby an embodied agent autonomously discovers the group structure of its action space through u