BriefGPT.xyz
May, 2023
神经傅里叶变换：一种等变表征学习的通用方法
Neural Fourier Transform: A General Approach to Equivariant Representation Learning
HTML
PDF
Masanori Koyama, Kenji Fukumizu, Kohei Hayashi, Takeru Miyato
TL;DR
提出了神经傅里叶变换（NFT）的概念，它是一种无需显式知道组在数据上的作用方式就可以学习组的潜在线性作用的通用框架。论文通过实验结果展示了NFT在不同场景下的应用。
Abstract
symmetry learning
has proven to be an effective approach for extracting the hidden structure of data, with the concept of
equivariance relation
playing the central role. However, most of the current studies are b
→