1986

David Rumelhart

1942–2011

Learning Internal Representations

With Geoffrey Hinton and Ronald Williams, published the influential backpropagation paper that helped revive multilayer neural networks.

Their 1986 Nature paper, Learning Representations by Back-propagating Errors, demonstrated how error signals could train useful internal representations across layers.

Backpropagation had earlier antecedents; this paper helped popularise its application to neural learning. Rumelhart's wider work connected computation, cognition and parallel distributed processing.

Dated contributions

  1. 1986

    Learning Internal Representations

    David Rumelhart, Geoffrey Hinton & Ronald J. Williams

    With Geoffrey Hinton and Ronald Williams, published the influential backpropagation paper that helped revive multilayer neural networks.

    Paper · 1986 — Learning representations by back-propagating errors ↗

Reference

Original work & historical records

Wikipedia