Distributional Hamilton-Jacobi-Bellman Equations for Continuous-Time Reinforcement Learning.
Harley E. WiltzerDavid MegerMarc G. BellemarePublished in: ICML (2022)
Keyphrases
- hamilton jacobi bellman
- optimal control
- reinforcement learning
- control problems
- dynamic programming
- approximate dynamic programming
- stochastic control
- infinite horizon
- control strategy
- state space
- control law
- rl algorithms
- function approximation
- model free
- learning algorithm
- nonlinear systems
- reinforcement learning algorithms
- brownian motion
- adaptive control
- average cost
- machine learning
- markov chain
- markov decision processes
- learning problems
- dynamical systems
- queueing systems