Analysis of stochastic gradient descent in continuous time

HIGHLIGHTS

• who: Jonas Latz from the (UNIVERSITY) have published the research work: Analysis of stochastic gradient descent in continuous time, in the Journal: (JOURNAL)
• what: The authors show that it converges weakly to the gradient flow with respect to the full target function as the learning rate approaches zero. The authors show that SGP is a sensible continuum limit of SGD and discuss SGP from a biological viewpoint: a model of the same type is used to model growth and phenotypes of clonal populations living in randomly fluctuating environments. In the following, the authors aim to . . .
  

   

  

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

  Thanks :)

  Username or E-mail:

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