当前位置: X-MOL 学术Found. Comput. Math. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Kähler Geometry of Framed Quiver Moduli and Machine Learning
Foundations of Computational Mathematics ( IF 3 ) Pub Date : 2022-08-01 , DOI: 10.1007/s10208-022-09587-3
George Jeffreys , Siu-Cheong Lau

We develop an algebro-geometric formulation for neural networks in machine learning using the moduli space of framed quiver representations. We find natural Hermitian metrics on the universal bundles over the moduli whose expressions are independent of dimension vector, and show that their Ricci curvatures give a Kähler metric on the moduli. Moreover, we use toric moment maps to construct activation functions and prove the universal approximation theorem for the softmax function (also known as Boltzmann distribution) using toric geometry of the complex projective space.



中文翻译:

框架箭袋模量和机器学习的 Kähler 几何

我们使用框架箭袋表示的模空间为机器学习中的神经网络开发代数几何公式。我们在模量上的通用束上找到自然 Hermitian 度量,其表达式与维向量无关,并表明它们的 Ricci 曲率给出了模量上的 Kähler 度量。此外,我们使用复曲面矩图构造激活函数,并使用复投影空间的复曲面几何证明 softmax 函数(也称为玻尔兹曼分布)的通用逼近定理。

更新日期:2022-08-02
down
wechat
bug