Evolutionary approach of reward function for reinforcement learning using genetic programming.
Shota SuminoAtsuko MutohShohei KatoPublished in: MHS (2011)
Keyphrases
- reward function
- reinforcement learning
- reinforcement learning algorithms
- markov decision processes
- state space
- optimal policy
- partially observable
- policy search
- inverse reinforcement learning
- markov decision process
- multiple agents
- transition model
- hierarchical reinforcement learning
- initially unknown
- transition probabilities
- function approximation
- machine learning
- dynamic programming
- learning agent
- learning algorithm
- temporal difference
- model free
- state action
- function approximators
- multi agent
- higher order
- markov decision problems
- control policies
- control policy
- action space
- policy iteration
- random walk
- complex systems
- action selection