Machine learningReinforcement learning
Q学习
Q学习由Christopher Watkins和Peter Dayan于1992年提出，是一种无模型强化学习算法，它仅凭经验学习在每个状态下采取每个动作的价值——即Q函数——而无需环境模型。它是离策略的：它在遵循探索性行为策略的同时学习最优动作价值，并在标准条件下可证明收敛于最优策略。
阅读完整方法
仅限会员
登录使用免费账户登录即可阅读本节。
Method map
The neighbourhood of related methods — select a node to explore.
来源
- Watkins, C. J. C. H., & Dayan, P. (1992). Q-learning. Machine Learning, 8(3–4), 279–292. DOI: 10.1007/BF00992698 ↗
- Sutton, R. S., & Barto, A. G. (2018). Reinforcement Learning: An Introduction (2nd ed.). MIT Press. ISBN: 978-0-262-03924-6
如何引用本页
ScholarGate. (2026, June 2). Q-Learning (Off-Policy Temporal-Difference Control). ScholarGate. https://scholargate.app/zh/machine-learning/q-learning
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
Compare side by side →