Amortized Score-Hamiltonian Policy Iteration: A Grid-Free Scheme for Relaxed Stochastic Control Problems
An amortized, grid-free implementation of continuous Langevin dynamics based policy-value iteration for entropy-regularized, infinite-horizon relaxed stochastic control problems and provides numerical experiments on general models to demonstrate the coupled actor--critic learning in high-dimensions.
Qin-Ai Feng, Wang Gu
· 1 citation