A mirror-descent method for deterministic optimal control achieves O(1/n) convergence under convexity assumptions and geometric rates with positive regularization, via relative smoothness and convexity estimates derived from Pontryagin's principle.
arXiv preprint arXiv:2506.02564 , year=
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
-
Mirror Descent for Deterministic Optimal Control
A mirror-descent method for deterministic optimal control achieves O(1/n) convergence under convexity assumptions and geometric rates with positive regularization, via relative smoothness and convexity estimates derived from Pontryagin's principle.