A penalty scheme and policy iteration for nonlocal HJB variational inequalities with monotone nonlinearities.
Christoph ReisingerYufei ZhangPublished in: Comput. Math. Appl. (2021)
Keyphrases
- variational inequalities
- policy iteration
- fixed point
- optimal control
- markov decision processes
- sensitivity analysis
- nonlinear programming
- infinite horizon
- least squares
- approximate dynamic programming
- average reward
- convex sets
- primal dual
- model free
- control problems
- reinforcement learning
- optimal policy
- sufficient conditions
- upper bound
- dynamic programming
- mathematical model
- markov random field
- temporal difference
- evolutionary algorithm
- objective function