Research breakthrough · curated entry
Watkins and Dayan publish Q-learning
Q-learning
- Date
- May 1992
- Category
- Research breakthrough
- Organization / Model
- Christopher Watkins and Peter Dayan
- Evidence
- 1 source · 1 evidence family
In the journal Machine Learning they proposed a model-free temporal-difference control method: the learner acquires an action-value function directly from delayed rewards and, under stated conditions, converges to the optimal policy. It became the algorithmic base of DQN and later deep reinforcement learning.
Curator's note: Cross-checked against the journal DOI; Q-learning first appeared in Watkins's 1989 doctoral thesis, and the 1992 paper gave the convergence proof.
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. Watkins 与 Dayan 发表 Q-learning，无模型强化学习奠基[EB/OL]. (1992-05)[2026-11-03]. https://aihist.net/e/q-learning-1992.html.
APA 7 · American Psychological Association
AI 纪年. (1992, May). Watkins 与 Dayan 发表 Q-learning，无模型强化学习奠基 [Web page]. https://aihist.net/e/q-learning-1992.html
MLA 9 · Modern Language Association
"Watkins 与 Dayan 发表 Q-learning，无模型强化学习奠基." AI 纪年, May 1992, aihist.net/e/q-learning-1992.html. Accessed 3 Nov. 2026.
Chicago 17 · notes-bibliography
AI 纪年. "Watkins 与 Dayan 发表 Q-learning，无模型强化学习奠基." Published May 1992. Accessed Nov. 3, 2026. https://aihist.net/e/q-learning-1992.html.
IEEE · citation-reference list
"Watkins 与 Dayan 发表 Q-learning，无模型强化学习奠基," AI 纪年, May 1992. [Online]. Available: https://aihist.net/e/q-learning-1992.html. [Accessed: 3 Nov. 2026].
ACM · author-date variant
AI 纪年. May 1992. Watkins 与 Dayan 发表 Q-learning，无模型强化学习奠基. Retrieved November 3, 2026 from https://aihist.net/e/q-learning-1992.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.