Mixture of von mises-fisher distribution with sparse prototypes

HIGHLIGHTS

• who: Fabrice Rossi and Florian Barbaro from the Dauphine-PSL, CEREMADE UMR, France have published the paper: Mixture of von Mises-Fisher distribution with sparse prototypes, in the Journal: (JOURNAL) of December/22,/2022
• what: The authors propose in this Article to estimate von Mises mixture using l1 penalized likelihood. The model`s behaviour studied simulated data and the authors show the advantages of the approach real data benchmark. The authors aim also at producing sparse directional means but the authors follow a different strategy. Following , the authors propose to use a l1 penalty for . . .
  

   

  

  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.