Research breakthrough · curated entry

LeCun trains a convolutional network on handwritten digits

Backpropagation Applied to Handwritten Zip Code Recognition

Date: June 1989
Category: Research breakthrough
Organization / Model: AT&T Bell Labs
Evidence: 1 source · 1 evidence family

LeCun and colleagues applied backpropagation to handwritten postal-code recognition, building one of the early convolutional networks. It showed that end-to-end training could work on a real image task.

Curator's note: Cross-checked against the author's publicly posted PDF.

cnn
Computer vision
People

Curated sources · 1

Official announcement · LeCun Publications1989-06-01

Backpropagation Applied to Handwritten Zip Code Recognition

yann.lecun.com

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. LeCun 用卷积网络识别手写数字[EB/OL]. (1989-06)[2026-11-03]. https://aihist.net/e/lenet-1989.html.

APA 7 · American Psychological Association

AI 纪年. (1989, June). LeCun 用卷积网络识别手写数字 [Web page]. https://aihist.net/e/lenet-1989.html

MLA 9 · Modern Language Association

"LeCun 用卷积网络识别手写数字." AI 纪年, June 1989, aihist.net/e/lenet-1989.html. Accessed 3 Nov. 2026.

Chicago 17 · notes-bibliography

AI 纪年. "LeCun 用卷积网络识别手写数字." Published June 1989. Accessed Nov. 3, 2026. https://aihist.net/e/lenet-1989.html.

IEEE · citation-reference list

"LeCun 用卷积网络识别手写数字," AI 纪年, June 1989. [Online]. Available: https://aihist.net/e/lenet-1989.html. [Accessed: 3 Nov. 2026].

ACM · author-date variant

AI 纪年. June 1989. LeCun 用卷积网络识别手写数字. Retrieved November 3, 2026 from https://aihist.net/e/lenet-1989.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.