Material Insights · Understanding the Material World

Backpropagation and gradient learning: Technologies

Technologies

The technologies that rely on it, and how each one uses it

Relied on by
  • DL Deep learning Deep networks are normally trained by passing the output error backwards to adjust the weights of every layer, the method set out in the 1986 paper [1].
  • LLM Large language models The models' weights are trained by gradient methods whose gradients are found by backpropagation.
  • AIx AI diagnosis from medical images Networks learn from graded images by adjusting every weight to reduce the error on each example.

as of 2026-11-12

Sources

  1. Learning representations by back-propagating errors Nature (D. E. Rumelhart, G. E. Hinton and R. J. Williams), via the SAO/NASA Astrophysics Data System, 1986, vol. 323, October 1986