True Online TD(λ)




Sarsa(λ)






Variable λ and γ


Watkins’s Q(λ) to Tree-Backup(λ)



Stable Off-policy Methods with Traces




Implementation Issues

实例: Sarsa(λ) on Mountain Car


代码编写:
调库

Tile coding


参数设置、定义


Sarsa(λ)



play Mountain Car for one episode

λ和α对Sarsa(λ)的影响

Sarsa(λ)算法的各版本比较

结果:



小结



