Deep Reinforcement Learning for Scheduling and Offloading in UAV-Assisted Mobile Edge Networks

Algorithm 1

traverse relaxation vectors and to find optimal and .

Input: Parameters , and

Output: ,

1: Elements in the vectors and are rearranged in ascending order to get and

2: Initialize as a very large number and and as empty vectors;

3: fordo

4: fordo

5: fordo

6: Compute Q() from (8);

7:   ifQ() < then

8:   ;

9:   [k] = ;

10:   ;

11:   end if

12:   end for

13:  end for

14: end for

15: return,.