Skip to main content

Table 2 Comparison of testing results with wm=0.5 for all mobile users

From: Decentralized computation offloading for multi-user mobile edge computing: a deep reinforcement learning approach

  Average reward Average power (watt) Average delay (ms)
  User 1 User 2 User 3 User 1 User 2 User 3 User 1 User 2 User 3
DDPG − 1.770 − 5.670 − 12.782 0.205 0.774 1.939 1.489 3.600 6.174
DQN − 2.174 − 7.657 − 14.688 0.292 1.156 2.320 1.428 3.753 6.176
GD-Offload − 2.514 − 7.597 − 18.690 0.402 1.309 3.143 1.007 2.103 5.951
GD-Local − 1.633 − 9.504 − 20.071 0.216 1.678 3.407 1.106 2.228 6.072