Adversarial retraining attack of asynchronous advantage actor-critic based pathfinding.
Tong ChenJiqiang LiuYingxiao XiangWenjia NiuEndong TongShuoru WangHe LiLiang ChangGang LiQi Alfred ChenPublished in: Int. J. Intell. Syst. (2021)
Keyphrases
- path finding
- actor critic
- path planning
- reinforcement learning
- policy gradient
- heuristic search
- search algorithm
- temporal difference
- approximate dynamic programming
- optimal control
- single agent
- hill climbing
- function approximation
- multi agent
- gradient method
- reinforcement learning algorithms
- rule learning
- neuro fuzzy
- optimal path
- machine learning
- average reward
- dynamic programming
- complexity analysis
- search space
- control system