Shrink and eliminate: a study of post-training quantization and repeated operations elimination in rnn models

HIGHLIGHTS

• who: Nesma M. Rezk and colleagues from the Department of Applied Physics and Electronics, Umeå University, Umeå, Sweden have published the research: Shrink and Eliminate: A Study of Post-Training Quantization and Repeated Operations Elimination in RNN Models, in the Journal: Information 2022, 13, 176. of /2022/
• what: The authors show how to apply post-training on these models with a minimal increase in the error by skipping of selected paths. The authors show that the of activation vectors in RNNs to integer precision leads to considerable sparsity if the delta networks method is applied . . .
  

   

  

  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.