https://marovi.ai/index.php?action=history&feed=atom&title=Translations%3AStochastic_Gradient_Descent%2F3%2Fzh Translations:Stochastic Gradient Descent/3/zh - Revision history 2026-04-27T22:02:18Z Revision history for this page on the wiki MediaWiki 1.39.1 https://marovi.ai/index.php?title=Translations:Stochastic_Gradient_Descent/3/zh&diff=5426&oldid=prev DeployBot: Batch translate Stochastic Gradient Descent unit 3 → zh 2026-04-27T03:38:12Z <p>Batch translate Stochastic Gradient Descent unit 3 → zh</p> <p><b>New page</b></p><div>在经典的{{Term|gradient descent|梯度下降}}中，每次参数更新前都要在整个训练集上计算{{Term|loss function|损失函数}}的完整梯度。当数据集很大时，这种做法的代价高得难以承受。SGD 通过在每一步从单个随机选取的样本（或一个小的 &#039;&#039;&#039;{{Term|mini-batch}}&#039;&#039;&#039;）估计梯度来解决该问题，以较高噪声的估计换取每次迭代成本的大幅降低。</div> DeployBot