Model-Free Risk-Sensitive Reinforcement Learning.
Grégoire DelétangJordi Grau-MoyaMarkus KuneschTim GeneweinRob BrekelmansShane LeggPedro A. OrtegaPublished in: CoRR (2021)
Keyphrases
- risk sensitive
- model free
- reinforcement learning
- reinforcement learning algorithms
- function approximation
- policy iteration
- temporal difference
- markov decision processes
- state space
- transfer learning
- learning problems
- policy evaluation
- rl algorithms
- optimal policy
- optimal control
- learning process
- average reward
- decision making
- partially observable markov decision processes
- action selection
- dynamical systems