ResNet: residual learning enables very deep networks
Kaiming He and colleagues introduced residual connections, allowing networks with 152+ layers to train; ResNet won ILSVRC-2015 with 3.57% top-5 error.
Key facts
- arXiv 1512.03385 (December 2015); CVPR 2016 best paper
- ILSVRC-2015 classification winner, 3.57% top-5 error
- Skip/residual connections are used in virtually all modern architectures, including Transformers
- Among the most-cited papers in all of science
What happened
Residual blocks learn a correction to an identity mapping, making optimization of very deep networks tractable.
Why it matters
Residual connections are a universal ingredient of deep learning; every Transformer block uses them.
Changelog
- 2026-09-29: created
Related events
- AlexNet wins ImageNet challenge, igniting the deep learning boom ★★★★★
- 'Attention Is All You Need' introduces the Transformer ★★★★★
Sources (2)
- paperDeep Residual Learning for Image Recognition (arXiv)
- discussionWikipedia: Residual neural network
id: 2015-12-10-resnet · updated 2026-09-29 · open in the interactive timeline