Traffic Status Prediction of Arterial Roads Based on the Deep Recurrent Q-Learning

Table 8

Parameter combinations for the second test.

Group	Learning rate	Reward delay	Greedy	Optimal loss coefficient
c	0.01	0.9	0.3	0.067
d	0.01	0.6	0.3	0.069
e	0.01	0.3	0.3	0.070