Research breakthrough · curated entry
ResNet introduces residual learning
Deep Residual Learning for Image Recognition
- Date
- December 10, 2015
- Category
- Research breakthrough
- Organization / Model
- Microsoft Research
- Evidence
- 1 source · 1 evidence family
He and colleagues proposed the residual network, using identity mappings to make networks a hundred layers deep trainable, and broke records on ImageNet. Residual connections have since become a standard component of almost every deep architecture.
Curator's note: Cross-checked against the arXiv paper.
Curated sources · 1
Cite this page
This page is a curated static snapshot and can be cited directly. Preferred format GB/T 7714—2025 (electronic resource [EB/OL]); the other 5 styles open below, together with the machine-readable files.
GB/T 7714—2025 (electronic resource [EB/OL])
AI 纪年. ResNet 提出残差学习，百层网络可训练[EB/OL]. (2015-12-10)[2026-11-03]. https://aihist.net/e/resnet-2015.html.
APA 7 · American Psychological Association
AI 纪年. (2015, December 10). ResNet 提出残差学习，百层网络可训练 [Web page]. https://aihist.net/e/resnet-2015.html
MLA 9 · Modern Language Association
"ResNet 提出残差学习，百层网络可训练." AI 纪年, 10 Dec. 2015, aihist.net/e/resnet-2015.html. Accessed 3 Nov. 2026.
Chicago 17 · notes-bibliography
AI 纪年. "ResNet 提出残差学习，百层网络可训练." Published Dec. 10, 2015. Accessed Nov. 3, 2026. https://aihist.net/e/resnet-2015.html.
IEEE · citation-reference list
"ResNet 提出残差学习，百层网络可训练," AI 纪年, 10 Dec. 2015. [Online]. Available: https://aihist.net/e/resnet-2015.html. [Accessed: 3 Nov. 2026].
ACM · author-date variant
AI 纪年. December 10, 2015. ResNet 提出残差学习，百层网络可训练. Retrieved November 3, 2026 from https://aihist.net/e/resnet-2015.html.
All six styles come from one template and can be compared word for word. [引用日期], Accessed, urldate and Y2 all carry this page's build date, 2026-11-03 — a static page is an immutable snapshot, so substitute the date you actually consulted it. The address in every style is the archive's canonical page: this English page translates the same snapshot, and the work keeps its name, so one record cites both. For one-click copying use the same entry in the timeline, which takes the day you opened it. Expand the other styles before printing.