Establishes Riemannian gradient flow equivalence for neural MMS steps, linear convergence under convexity conditions, and O(δ) tracking bounds for inexact iterates.
An introduction to the analysis of gradients systems.arXiv preprint arXiv:2306.05026,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Global Convergence and Error Propagation in Neural Gradient Flows: A Riemannian Optimization Framework
Establishes Riemannian gradient flow equivalence for neural MMS steps, linear convergence under convexity conditions, and O(δ) tracking bounds for inexact iterates.