Iteration (machine learning)

Last revised by Andrew Murphy on 2 May 2019

An iteration is a term used in machine learning and indicates the number of times the algorithm's parameters are updated. Exactly what this means will be context dependent. A typical example of a single iteration of training of a neural network would include the following steps:

  1. processing the training dataset batch
  2. calculating the cost function
  3. backpropagation and adjustment of all weighting factors

Training of a neural network will require many iterations. 

Machine learning iteration is conceptually related to but not the same as other concepts in computer programming of iteration. The general meaning of iteration in image related programming is applying a function to every element of something. To iterate over an image is to apply a given function to every pixel or voxel value. Many simple functions such as windowing that radiologists execute are accomplished through functions that iterate over all the values in an image. In contrast to the meaning of iteration in image processing, Iteration in the context of machine learning refers to many processes involved in training the computer program.  

ADVERTISEMENT: Supporters see fewer/no ads

Updating… Please wait.

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

 Thank you for updating your details.