Asynchronous reinforcement learning algorithms for solving discrete space path planning problems.
Xingyu ZhaoShifei DingYuexuan AnWeikuan JiaPublished in: Appl. Intell. (2018)
Keyphrases
- planning problems
- reinforcement learning algorithms
- state space
- discrete space
- solving planning problems
- reinforcement learning
- heuristic search
- markov decision processes
- markov decision problems
- dynamic programming
- domain independent
- euclidean space
- learning algorithm
- optimal policy
- markov chain
- state variables
- model free
- temporal difference
- knapsack problem
- continuous state
- markov decision process
- particle filter
- planning domains
- multi agent
- learning agent
- data points
- supervised learning
- higher order
- classical planning
- closed curves
- reward function
- function approximation
- search space