Stochastic Optimization Under Power-Law Spectra: Tight Bounds and Shuffling Analysis
arXiv:2609.36271v1 Announce Type: cross Abstract: Recent work has established that power-law spectral conditions on data enable tight convergence bounds for deterministic gradient descent, resolving the conflict between classical exponential bounds and observed power-law learning curves. In this wo…
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-09-30 04:00 · arXiv stat.ML
Stochastic Optimization Under Power-Law Spectra: Tight Bounds and Shuffling Analysis