Backpropagation (machine learning)

Andrew Murphy and Dr Matt Adams et al.

Backpropagation in supervised machine learning is the process used to calculate the gradient associated with each parameter weighting. Essentially, the gradient estimates how the system parameters should change in order to optimize the network overall 1,2.

Share article

Article information

rID: 56164
Synonyms or Alternate Spellings:

Support Radiopaedia and see fewer ads

Updating… Please wait.

 Unable to process the form. Check for errors and try again.

 Thank you for updating your details.