Deep learning experiments in Cohen et al. (2021) using deterministic Gradient Descent (GD) revealed an {\em Edge of Stability (EoS)} phase when learning rate (LR) and sharpness (\emph{i.e.}, the largest eigenvalue of Hessian) no longer behave as in traditional optimization. Sharpness stabilizes around $2/$LR and loss goes up and down across iterations, yet still with an overall downward trend. The current paper mathematically analyzes a new mechanism of implicit regularization in the EoS phase, whereby GD updates due to non-smooth loss landscape turn out to evolve along some deterministic flow on the manifold of minimum loss. This is in contrast to many previous results about implicit bias either relying on infinitesimal updates or noise in gradient. Formally, for any smooth function $L$ with certain regularity condition, this effect is demonstrated for (1) {\em Normalized GD}, i.e., GD with a varying LR $ \eta_t =\frac{ \eta }{ || \nabla L(x(t)) || } $ and loss $L$; (2) GD with constant LR and loss $\sqrt{L}$. Both provably enter the Edge of Stability, with the associated flow on the manifold minimizing $\lambda_{\max}(\nabla^2 L)$. The above theoretical results have been corroborated by an experimental study.

研究了神经网络训练中的难点问题Edge of Stability，发现了一种新的内隐正则化机制，通过对最小化损失面的低维流动，提出对比以往对无穷小更新或梯度噪声的依赖。

深度学习中稳定性边缘处的梯度下降理解