Deep Reinforcement Learning for UAV Intelligent Mission Planning

Table 5

Hyperparameter settings.

Parameter	Value
max_episode	300
learning_rate	0.0003
Clip	0.2
Gamma	0.995
Lambda	0.95
max_step_per_round	1000
coeff_loss_policy	1
coeff_loss_value	0.5
coeff_loss_entropy	0.01
minibatch_size	64
num_seed	3