Risk-Sensitive Reinforcement Learning: Iterated CVaR and the Worst Path.
Yihan DuSiwei WangLongbo HuangPublished in: CoRR (2022)
Keyphrases
- risk sensitive
- reinforcement learning
- model free
- optimal control
- markov decision processes
- risk neutral
- control policies
- markov decision problems
- reinforcement learning algorithms
- utility function
- function approximation
- state space
- dynamic programming
- risk averse
- machine learning
- temporal difference
- finite state
- optimal policy
- expected utility
- control strategies
- partially observable
- policy iteration
- action space
- robust optimization
- control strategy
- average cost
- finite horizon
- portfolio selection
- multistage
- efficient optimization
- learning algorithm