BriefGPT.xyz
May, 2017
矩阵分解中的隐式正则化
Implicit Regularization in Matrix Factorization
HTML
PDF
Suriya Gunasekar, Blake Woodworth, Srinadh Bhojanapalli, Behnam Neyshabur, Nathan Srebro
TL;DR
通过使用矩阵因式分解的梯度下降法来优化欠定二次目标函数时，对步长采用合适大小以及初始值足够接近原点进行隐式正则化会使得梯度下降法收敛到最小核范数解，这一结论在实证和理论方面都得到了支持。
Abstract
We study
implicit regularization
when optimizing an
underdetermined quadratic objective
over a matrix $X$ with
gradient descent
on a facto
→