Fully fourier space spherical convolutional neural network based on Clebsch-Gordan transforms
US11934478B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Jun 20, 2019 |
| Grant date | Mar 19, 2024 |
| Priority date | — |
| Expiry date | Nov 29, 2040 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T15/08
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Methods and systems for computationally processing data with a multi-layer convolutional neural network (CNN) having an input and output layer, and one or more intermediate layers are described. Input data represented in a form of evaluations of continuous functions on a sphere may be received at a computing device and input to the input layer. The input layer may compute outputs as covariant Fourier space activations by transforming the continuous functions into spherical harmonic expansions. The output activations from the input layer may be processed sequentially through each of the intermediate layers. Each, intermediate layer may apply Ciebsch-Gordan transforms to compute respective covariant Fourier space activations as input to an immediately next layer, without computing any intermediate inverse Fourier transforms or forward Fourier transforms. Finally, the respective covariant Fourier space activations of the last intermediate layer may be processed in the output layer of the CNN to compute invariant activations.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.