Constructing Invariant and Equivariant Operations by Symmetric Tensor Network
This work addresses the need for efficient symmetry-aware operations in geometric deep learning, with applications in geometry graph neural networks and material constitutive law learning, representing a novel method for a known bottleneck.
The paper tackles the problem of systematically constructing invariant and equivariant operations for neural networks that incorporate symmetry, presenting a method that handles various tensor types and includes a graphical representation using symmetric tensor networks to simplify proofs and constructions.
Design of neural networks that incorporate symmetry is crucial for geometric deep learning. Central to this effort is the development of invariant and equivariant operations. This works presents a systematic method for constructing valid invariant and equivariant operations. It can handle inputs and outputs in the form of Cartesian tensors with different rank, as well as spherical tensors with different types. In addition, our method features a graphical representation utilizing the symmetric tensor network, which simplifies both the proofs and constructions related to invariant and equivariant functions. We also apply this approach to design the equivariant interaction message for the geometry graph neural network, and equivariant machine learning model to learn the constitutive law of materials.