Backward Q-learning: The combination of Sarsa algorithm and Q-learning.
Yin-Hao WangTzuu-Hseng S. LiChih-Jui LinPublished in: Eng. Appl. Artif. Intell. (2013)
Keyphrases
- learning algorithm
- reinforcement learning
- function approximation
- dynamic programming
- bucket brigade
- cooperative
- temporal difference learning
- model free
- stochastic approximation
- optimal solution
- computational complexity
- path planning
- matching algorithm
- state space
- reinforcement learning algorithms
- np hard
- cost function
- preprocessing
- multi agent
- objective function
- computational cost
- simulated annealing
- expectation maximization
- segmentation algorithm
- optimal policy
- markov decision processes
- significant improvement
- k means
- actor critic
- td learning
- similarity measure