Jan 18, 2024 · We introduce the Gaunt Tensor Product, which serves as a new method to construct efficient equivariant operations across different model architectures.

Dec 7, 2023 · The key idea is to convert the Clebasch-Gordan coefficients to Gaunt coefficients, by which the tensor product of irreps can be calculated ...

Jan 19, 2024 · This perspective further allows us to change the basis for the equivariant operations from spherical harmonics to a 2D Fourier basis.

In this work, we propose a systematic approach to substantially accelerate the computation of the tensor products of irreps. We mathematically connect the ...

We now introduce our Gaunt Tensor Product, which serves as a new method for efficient E(3)- equivariant operations.

Jan 18, 2024 · The Gaunt Tensor Product is introduced, which serves as a new method to construct efficient equivariant operations across different model architectures.

Jan 18, 2024 · The Gaunt Tensor Product is introduced as a computational technique that speeds up equivariant operations using a Fourier basis and Fast Fourier ...

Jan 18, 2024 · In thiswork, we propose a systematic approach to substantially accelerate thecomputation of the tensor products of irreps. We mathematically ...

2 days ago · Developing equivariant neural networks for the E(3) group plays an important role in modeling 3D data across real-world applications. Enforcing ...

Jan 20, 2024 · The proposed Gaunt tensor product serves as an efficient way to compute equivariant operations like feature interactions, convolutions, and many-body ...