Solving Problems with Extended Reachability Goals through Reinforcement Learning on Propositionally Constrained State Spaces.
Anderson V. de AraújoCarlos H. C. RibeiroPublished in: SMC (2013)
Keyphrases
- solving problems
- state space
- reinforcement learning
- reinforcement learning algorithms
- optimal policy
- markov decision processes
- orders of magnitude
- learning agent
- action space
- markov chain
- function approximation
- continuous state spaces
- dynamic programming
- heuristic search
- partially observable
- planning problems
- markov decision process
- model free
- state abstraction
- particle filter
- learning algorithm
- dynamical systems
- search space
- learning problems
- multi agent
- machine learning
- optimal control
- action selection
- belief state
- control problems
- partially observable markov decision processes
- interior point methods
- data sets