Research on Vibration Reduction Control Based on Reinforcement Learning

Algorithm 1

The updating method of value function for the different reinforcement learning algorithms.

	Input: environment E; action space A; initial state ; discount coefficient ; learning rate ;
(1)
(2)
(3)	For
(4)	reward and transfer status generated by performing actions in environment E;
(5)
(6)
(7)
(8)
(9)	End
	Output: strategy