arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.06964cs.LGcs.AI
收藏

损失景观中存在一个奇点

There is a Singularity in the Loss Landscape

Mark Lowell

AI总结:

本文通过实验发现,随着数据集增大,损失景观中出现梯度无界的奇点,梯度下降使网络靠近该奇点,并解释了稳定性边缘训练等Hessian相关现象,且顶部子空间虽占梯度大部分但对学习贡献小。

摘要:

尽管神经网络被广泛采用,但其训练动态仍然鲜为人知。我们通过实验表明,随着数据集规模的增大,会出现一个点,在该点处损失梯度的幅值变得无界。梯度下降迅速将网络拉近参数空间中的这个奇点,后续训练在其附近进行。这个奇点解释了最近在神经网络损失函数的Hessian矩阵中观察到的多种现象,例如在稳定性边缘训练以及梯度在顶部子空间中的集中。一旦网络接近该奇点,顶部子空间对学习的贡献就很小,尽管它构成了梯度的大部分。

英文摘要:

Despite the widespread adoption of neural networks, their training dynamics remain poorly understood. We show experimentally that as the size of the dataset increases, a point forms where the magnitude of the gradient of the loss becomes unbounded. Gradient descent rapidly brings the network close to this singularity in parameter space, and further training takes place near it. This singularity explains a variety of phenomena recently observed in the Hessian of neural network loss functions, such as training on the edge of stability and the concentration of the gradient in a top subspace. Once the network approaches the singularity, the top subspace contributes little to learning, even though it constitutes the majority of the gradient.

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。

↑