Prehistory — From Neural Networks to the TransformerSuperVision (Krizhevsky, Sutskever, Hinton, University of Toronto)Ilya SutskeverGeoffrey HintonUniversity of Toronto

SuperVision wins ILSVRC 2012 with a 15.3% top-5 error rate

Won ILSVRC 2012 with a top-5 error rate of 15.3%; the results were announced at the ECCV 2012 workshop

Why it matters

The turning point at which, with a wide margin over second place, deep CNNs became the mainstream methodology of computer vision

Source — Primary sources first; secondary reports are labelled “Press”

Officialimage-net.orghttps://www.image-net.org/challenges/LSVRC/2012/results.html

Added in Week 35, 2026 · See that week

Nearby on the same axis

ImageNet Classification with Deep Convolutional Neural Networks (AlexNet)PaperImproving neural networks by preventing co-adaptation of feature detectorsPaperRepresentation Learning: A Review and New PerspectivesPaperDeep Neural Networks for Acoustic Modeling in Speech RecognitionOfficial