Establishes linear L2 convergence of dual gradient descent for quadratically regularized OT via spectral analysis showing the linearized operator is a strict contraction.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
math.OC 2verdicts
UNVERDICTED 2representative citing papers
Proves local error bound and PL inequality for QOT dual with explicit constants, enabling linear convergence of ascent methods.
citing papers explorer
-
Linear Convergence of Gradient Descent for Quadratically Regularized Optimal Transport
Establishes linear L2 convergence of dual gradient descent for quadratically regularized OT via spectral analysis showing the linearized operator is a strict contraction.
-
Polyak-Lojasiewicz Inequality for Quadratically Regularized Optimal Transport
Proves local error bound and PL inequality for QOT dual with explicit constants, enabling linear convergence of ascent methods.