Large Spikes in Stochastic Gradient Descent: A Large-Deviations View

arXiv:2603.10079v2 Announce Type: replace Abstract: Large loss spikes in stochastic gradient descent are studied through a rigorous large-deviations analysis for a shallow, fully connected network in the NTK scaling. In contrast to full-batch gradient descent, the catapult phase is shown to split into inflationary and deflationary regimes, determined by an explicit log-drift criterion. In both cases, large spikes are shown to be at least polynomially likely. In addition, these spikes are shown to be the dominant mechanism by which sharp minima are escaped and curvature is reduced, thereby favouring flatter solutions. Corresponding results are also obtained for certain ReLU networks, and implications for curriculum learning are derived.

Fisher-Geometric Diffusion in Stochastic Gradient Descent: Optimal Rates, Oracle Complexity, and Information-Theoretic Limits
Product-Stability: Provable Convergence for Gradient Descent on the Edge of Stability
Stochastic Gradient Descent in the Saddle-to-Saddle Regime of Deep Linear Networks
Non-Expansive Mappings in Two-Time-Scale Stochastic Approximation: Finite-Time Analysis
Optimal Rates for Generalization of Gradient Descent for Deep ReLU Classification

Source: arXiv cs.LG | 2026-04-14

Related