Model-Free Characterizations of the Hamilton-Jacobi-Bellman Equation and Convex Q-Learning in Continuous Time.
Fan LuJoel MathiasSean P. MeynKaranjit KalsiPublished in: CoRR (2022)
Keyphrases
- hamilton jacobi bellman
- model free
- optimal control
- reinforcement learning
- control problems
- reinforcement learning algorithms
- function approximation
- policy iteration
- dynamical systems
- stochastic control
- state space
- dynamic programming
- nonlinear systems
- temporal difference
- approximate dynamic programming
- policy evaluation
- rl algorithms
- brownian motion
- learning algorithm
- infinite horizon
- control strategy
- control law
- optimal policy
- learning agent
- markov processes
- machine learning
- fuzzy control
- control policy
- average reward
- queueing systems
- learning tasks
- transfer learning
- linear programming
- multi agent