Large loss spikes in SGD are polynomially likely and serve as the dominant mechanism for escaping sharp minima toward flatter solutions in the NTK regime.
The two regimes of deep network training
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Derives second-order path-kernel interpolation formulas for gradient descent, SGD, and momentum training, adding curvature terms and a concentration estimate around the expected prediction.
citing papers explorer
-
Large Spikes in Stochastic Gradient Descent: A Large-Deviations View
Large loss spikes in SGD are polynomially likely and serve as the dominant mechanism for escaping sharp minima toward flatter solutions in the NTK regime.
-
Second-Order Path Kernel Interpolation Formulas in Machine Learning
Derives second-order path-kernel interpolation formulas for gradient descent, SGD, and momentum training, adding curvature terms and a concentration estimate around the expected prediction.