Risk-Sensitive Reinforcement Learning.
Yun ShenMichael J. TobiaTobias SommerKlaus ObermayerPublished in: Neural Comput. (2014)
Keyphrases
- risk sensitive
- reinforcement learning
- model free
- markov decision processes
- optimal control
- control policies
- markov decision problems
- reinforcement learning algorithms
- optimal policy
- function approximation
- utility function
- state space
- policy iteration
- markov decision chains
- finite state
- dynamic programming
- temporal difference
- action space
- action selection
- partially observable
- reward function
- machine learning
- multi agent
- learning algorithm
- markov decision process
- average reward
- infinite horizon
- optimality criterion