Large Learning Rate Matters for Non-Convex Optimization
Lists
Only you see this
Abstract
When training neural networks, it has been widely observed that a large step size is essential in stochastic gradient descent (SGD) for obtaining superior models. However, the effect of large step sizes on the success of SGD is not well understood theoretically. Several previous works have attributed this success to the stochastic noise present in SGD. However, we show through a novel set of experiments that the stochastic noise is not sufficient to explain good non-convex training, and that instead the effect of a large learning rate itself is essential for obtaining best performance. We demonstrate the same effects also in the noise-less case, i.e. for full-batch GD. We formally prove that GD with large step size---on certain non-convex function classes---follows a different trajectory than GD with a small step size, which can lead to convergence to a global minimum instead of a local one. Finally, we also demonstrate the difference in trajectories for small and large learning rates for real neural networks, again observing that large learning rates allow escaping from a local minimum, confirming this behavior is indeed relevant in practice.
BibTeX
@inproceedings{mohtashami2023large,
author = {Amirkeivan Mohtashami and Martin Jaggi and Sebastian U Stich},
title = {{Large Learning Rate Matters for Non-Convex Optimization}},
booktitle = {International Conference on Learning Representations},
year = {2023},
url = {https://openreview.net/forum?id=JsrvkgM8gO2},
}Loading related content...