Policy Gradient Methods for Risk-Sensitive Distributional Reinforcement Learning with Provable Convergence.
Minheng XiaoXian YuLei YingPublished in: CoRR (2024)
Keyphrases
- risk sensitive
- policy gradient methods
- optimal control
- reinforcement learning
- actor critic
- model free
- policy gradient
- markov decision processes
- natural actor critic
- markov decision problems
- control policies
- reinforcement learning algorithms
- reinforcement learning problems
- function approximation
- dynamic programming
- utility function
- function approximators
- policy iteration
- optimal policy
- reinforcement learning methods
- temporal difference
- partially observable
- control strategy
- control strategies
- average cost
- average reward
- convergence rate
- state space
- multi agent
- finite state
- least squares
- control policy
- partially observable markov decision processes
- reward function