arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1806.01337stat.MLcs.LG
收藏

Backdrop:随机反向传播

Backdrop: Stochastic Backpropagation

New York University(纽约大学)

Siavash Golkar, Kyle Cranmer

摘要:

我们介绍了background,一个灵活且易于实现的方法,直观地描述为仅沿反向传播管道起作用的dropout。背景是通过一个或多个掩蔽层来实现的,这些掩蔽层沿着网络插入到特定点。每个背景掩蔽层在前向传递中充当身份,但随机掩蔽后向梯度传播的部分。直观地说,在任何卷积层之后插入背景层会导致对应于该尺度特征的随机梯度。因此,background非常适合数据具有多尺度、层次结构的问题。Backdrop也可以应用于标准SGD方法不太适合的不可分解损失函数的问题。我们进行了大量的实验,并证明背景导致显着改善泛化。

英文摘要:

We introduce backdrop, a flexible and simple-to-implement method, intuitively described as dropout acting only along the backpropagation pipeline. Backdrop is implemented via one or more masking layers which are inserted at specific points along the network. Each backdrop masking layer acts as the identity in the forward pass, but randomly masks parts of the backward gradient propagation. Intuitively, inserting a backdrop layer after any convolutional layer leads to stochastic gradients corresponding to features of that scale. Therefore, backdrop is well suited for problems in which the data have a multi-scale, hierarchical structure. Backdrop can also be applied to problems with non-decomposable loss functions where standard SGD methods are not well suited. We perform a number of experiments and demonstrate that backdrop leads to significant improvements in generalization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑