A note on factor normalization for deep neural network models

HIGHLIGHTS

• who: Haobo Qi from the Guanghua SchoolPeking University have published the research work: A note on factor normalization for deep neural network models, in the Journal: Scientific Reports Scientific Reports of 25/09/2021
• what: Because the SGD algorithm is a stochastic version of the gradient descent (GD) algorithm, the authors focus on the standard GD algorithm in this section for simplicity. To address this impor tant problem, the authors evaluate the expected Hessian matrix as H=E(Lu0308 (u03b8))=E{u2113u0308(Yi, Xiu22a4 u03b8)Xi Xiu22a4 }. To further explain the motivation behind the proposed . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

  If you don't have an account, you can create one here.