通过 Gaunt 张量积在傅里叶基中实现高效等变运算
Enabling Efficient Equivariant Operations in the Fourier Basis via Gaunt Tensor Products
AI总结:
本文提出 Gaunt 张量积方法,将等变运算从球谐基转换到二维傅里叶基,利用卷积定理和快速傅里叶变换将不可约表示张量积复杂度从 O(L^6) 降至 O(L^3),并在 Open Catalyst Project 和 3BPA 数据集上验证了效率与性能的提升。
摘要:
开发用于 E(3) 群的等变神经网络在现实应用中对三维数据建模起着重要作用。强制这种等变性主要涉及不可约表示(irreps)的张量积。然而,随着高阶张量的使用,此类运算的计算复杂度显著增加。在这项工作中,我们提出了一种系统方法,以大幅加速不可约表示张量积的计算。我们在数学上将常用的 Clebsch-Gordan 系数与 Gaunt 系数联系起来,后者是三个球谐函数乘积的积分。通过 Gaunt 系数,不可约表示的张量积等价于由球谐函数表示的球函数之间的乘法。这一视角进一步使我们能够将等变运算的基从球谐函数更改为二维傅里叶基。因此,由二维傅里叶基表示的球函数之间的乘法可以通过卷积定理和快速傅里叶变换高效计算。这种变换将不可约表示的全张量积的复杂度从 $\mathcal{O}(L^6)$ 降低到 $\mathcal{O}(L^3)$,其中 $L$ 是不可约表示的最大度数。利用这种方法,我们引入了 Gaunt 张量积,作为一种在不同模型架构中构建高效等变运算的新方法。我们在 Open Catalyst Project 和 3BPA 数据集上的实验证明了我们方法的效率提升和性能改进。
英文摘要:
Developing equivariant neural networks for the E(3) group plays an important role in modeling 3D data across real-world applications. Enforcing this equivariance primarily involves the tensor products of irreducible representations (irreps). However, the computational complexity of such operations increases significantly as higher-order tensors are used. In this work, we propose a systematic approach to substantially accelerate the computation of the tensor products of irreps. We mathematically connect the commonly used Clebsch-Gordan coefficients to the Gaunt coefficients, which are integrals of products of three spherical harmonics. Through Gaunt coefficients, the tensor product of irreps becomes equivalent to the multiplication between spherical functions represented by spherical harmonics. This perspective further allows us to change the basis for the equivariant operations from spherical harmonics to a 2D Fourier basis. Consequently, the multiplication between spherical functions represented by a 2D Fourier basis can be efficiently computed via the convolution theorem and Fast Fourier Transforms. This transformation reduces the complexity of full tensor products of irreps from $\mathcal{O}(L^6)$ to $\mathcal{O}(L^3)$, where $L$ is the max degree of irreps. Leveraging this approach, we introduce the Gaunt Tensor Product, which serves as a new method to construct efficient equivariant operations across different model architectures. Our experiments on the Open Catalyst Project and 3BPA datasets demonstrate both the increased efficiency and improved performance of our approach.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。