Reduction of state space in reinforcement learning by sensor selection.
Yasutaka KishimaKentarou KurashigePublished in: Artif. Life Robotics (2013)
Keyphrases
- state space
- reinforcement learning
- reinforcement learning algorithms
- markov decision processes
- optimal policy
- partially observable
- sensor data
- markov decision process
- heuristic search
- real time
- function approximation
- dynamic programming
- action space
- dynamical systems
- sensor networks
- markov chain
- particle filter
- model free
- state variables
- temporal difference
- reward function
- machine learning
- continuous state spaces
- policy iteration
- state abstraction
- initial state
- policy search
- goal state
- optimal control
- selection algorithm
- multi sensor
- supervised learning
- search space
- data acquisition
- rough sets
- control policy
- temporal difference learning
- continuous state
- multi agent
- genetic algorithm
- reward shaping