Deep learning with dynamically weighted loss health management

HIGHLIGHTS

• who: Divish Rengasamy and collaborators from the Gas Turbine and Transmissions Research Centre, The University of Nottingham, Nottingham , RD, UK have published the paper: Deep Learning with Dynamically Weighted Loss Health Management, in the Journal: (JOURNAL)
• what: This paper demonstrated that the PHM of a gas turbine engine and APS system were improved by using deep learning models with a dynamically weighted loss function that focused on instances that were poorly learned during the training process.
• how: This approach modifies the loss function so that the weights associated with it are dynamically calculated . . .
  

   

  

  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.