Introduced by Hinton et al. in 2012, dropout has stood the test of time as a regularizer for preventing overfitting in neural networks. In this study, we demonstrate that dropout can also mitigate underfitting when used at the start of training. During the early phase, we find dropout reduces the directional variance of gradients across mini-batches and helps align the mini-batch gradients with the entire dataset's gradient. This helps counteract the stochasticity of SGD and limit the influence of individual batches on model training. Our findings lead us to a solution for improving performance in underfitting models - early dropout: dropout is applied only during the initial phases of training, and turned off afterwards. Models equipped with early dropout achieve lower final training loss compared to their counterparts without dropout. Additionally, we explore a symmetric technique for regularizing overfitting models - late dropout, where dropout is not used in the early iterations and is only activated later in training. Experiments on ImageNet and various vision tasks demonstrate that our methods consistently improve generalization accuracy. Our results encourage more research on understanding regularization in deep learning and our methods can be useful tools for future neural network training, especially in the era of large data. Code is available at https://github.com/facebookresearch/dropout.

翻译：Hinton等人于2012年提出的丢弃法（dropout）作为防止神经网络过拟合的正则化技术，经受了时间的考验。本研究表明，在训练初期使用丢弃法时，它同样能够缓解欠拟合。在早期阶段，我们发现丢弃法降低了跨小批量梯度的方向方差，并有助于使小批量梯度与整个数据集的梯度对齐。这有助于抵消SGD的随机性，并限制单个批次对模型训练的影响。我们的发现促使我们提出一种提升欠拟合模型性能的解决方案——早期丢弃法：仅在训练初始阶段应用丢弃法，随后将其关闭。与未使用丢弃法的模型相比，配备早期丢弃法的模型在最终训练损失上更低。此外，我们探索了一种用于正则化过拟合模型的对称技术——晚期丢弃法，即在训练早期迭代中不使用丢弃法，仅在后期激活。在ImageNet及多种视觉任务上的实验表明，我们的方法始终能提升泛化准确率。这一结果鼓励对深度学习正则化机制展开进一步研究，而我们的方法有望成为未来神经网络训练中的实用工具，尤其是在大数据时代。代码开源地址：https://github.com/facebookresearch/dropout。