随机梯度下降中隐式正则化的起源 Rogerspy 2021-04-06 论文解读 字数统计: 1.8k字 | 阅读时长≈ 6分 首先推荐两篇论文: Samuel L Smith, Benoit Dherin, David Barrett, Soham De (2021) On the Origin of Implicit Regularization in Stochastic Gradient Descent David G.T. Barrett, Benoit Dherin (2021) Implicit Gradient Regularization 阅读全文 NLP 隐式正则化