Machine learning
随机梯度下降 (SGD)
随机梯度下降 (SGD) 是一种一阶迭代优化算法，源于 Robbins 和 Monro 于 1951 年提出的随机逼近框架。它通过在每一步使用单个随机选取的训练样本（或一个小批量样本）计算出的梯度来更新模型参数，从而最小化目标函数。它是现代机器学习和深度学习的核心优化引擎，能够处理内存无法容纳的大型数据集的训练。
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
- Robbins, H. & Monro, S. (1951). A Stochastic Approximation Method. The Annals of Mathematical Statistics, 22(3), 400–407. DOI: 10.1214/aoms/1177729586 ↗
- Goodfellow, I., Bengio, Y. & Courville, A. (2016). Deep Learning (Ch. 8). MIT Press. ISBN: 978-0-262-03561-3
如何引用本页
ScholarGate. (2026, June 3). Stochastic Gradient Descent (SGD) Optimization Algorithm. ScholarGate. https://scholargate.app/zh/machine-learning/stochastic-gradient-descent
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 →