Non-normality in linearized optimizer update operators yields a pseudospectral bound where κ(V) warns of transient amplification before spectral radius indicates instability.
Hessian-based analysis of large batch training and robustness to adversaries.Advances in Neural Information Processing Systems, 31, 2018
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
Hessian eigenvector displacement and inverse participation ratio metrics show SGD stabilizing leading curvature directions while Adam causes more reorganization and parameter localization in MLP training.
citing papers explorer
-
Non-normal spectral signatures of instability in neural network training dynamics
Non-normality in linearized optimizer update operators yields a pseudospectral bound where κ(V) warns of transient amplification before spectral radius indicates instability.
-
Characterizing Optimizer-Dependent Training Dynamics Through Hessian Eigenvector Displacement and Localization
Hessian eigenvector displacement and inverse participation ratio metrics show SGD stabilizing leading curvature directions while Adam causes more reorganization and parameter localization in MLP training.