For linear state dynamics with control-dependent diffusion, proximal policy gradient iterates converge linearly to a stationary control when the running or terminal cost is sufficiently strongly convex.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Convergence of Proximal Policy Gradient Method for Problems with Control Dependent Diffusion Coefficients
For linear state dynamics with control-dependent diffusion, proximal policy gradient iterates converge linearly to a stationary control when the running or terminal cost is sufficiently strongly convex.