A backstepping control framework for continuous-time unconstrained optimization that recovers Nesterov flow and PID accelerated optimizers as special cases for convex functions.
Con- strained optimization from a control perspective via feedback lineariza- tion.arXiv preprint arXiv:2503.12665, 2025
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
fields
math.OC 4representative citing papers
A PID feedback law on dual variables induces a unified family of saddle-point flows for constrained optimization, with explicit global exponential convergence guarantees under convexity and affine constraints.
Proves global exponential convergence of PI and feedback linearization Lagrangian flows for non-convex equality-constrained optimization under a manifold-restricted convexity property.