A scaled gradient-momentum framework achieves global finite-time convergence by linking gradient-dominance properties of the objective to finite-time stability via state-dependent scaling.
Accelerated optimization in deep learning with a proportional-integral-derivative controller,
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
-
Finite-Time Optimization via Scaled Gradient-Momentum Flows
A scaled gradient-momentum framework achieves global finite-time convergence by linking gradient-dominance properties of the objective to finite-time stability via state-dependent scaling.