Learning Symmetric Embeddings for Equivariant World Models

Understand

Incorporating symmetries can lead to highly data-efficient and generalizable models by defining equivalence classes of data samples related by transformations.

• However, characterizing how transformations act on input data is often difficult, limiting the applicability of equivariant models.
• We propose learning symmetric embedding networks (SENs) that encode an input space (e.g.
• images), where we do not know the effect of transformations (e.g.