Computation offloading strategy based on deep reinforcement learning for connected and autonomous vehicle in vehicular edge computing

Journal of Cloud Computing

Advances, Systems and Applications

Table 3 Parameters setting about RL algorithms

Description	Parameter	Value
Learning rate	α	0.01
Discount factor	γ	0.9
Trace decay rate	λ	0.5
Initial temperature	θ	0.9
Number of hidden layers		2
Number of nodes for the first hidden layer		20
Number of nodes for the second hidden layer		20
Activation function for hidden layers		ReLU
Maximum replay memory size	\|D\|	500
Minibatch size		300
Parameter updating frequency for target DQN	ι	10