Backpropagation
Rumelhart, Hinton and Williams show how a multi-layer network can correct its errors by propagating the error signal backwards. This helps overcome the limitation highlighted by Minsky and Papert.
Room 4, 1986–2024
Neural networks
An artificial neuron does something simple: it weighs its inputs and produces a decision. The power comes from connecting thousands or billions of such units in layers and adjusting their weights through training.
Open the interactive room →Exhibits in this room
A network that reads
Yann LeCun trains a network to recognise handwritten digits in postal ZIP codes. Related systems were later used for tasks such as reading bank cheques.
AlexNet
A deep neural network trained on graphics processors wins the ImageNet image-recognition competition by a large margin. The modern deep-learning era accelerates.
AlphaGo
DeepMind’s system defeats Lee Sedol, one of the world’s leading Go players, in a game long considered exceptionally difficult for machines.
Nobel Prize in Physics
John Hopfield and Geoffrey Hinton are awarded the 2024 Nobel Prize in Physics for foundational discoveries and inventions that enabled machine learning with artificial neural networks.