Graph Reinforcement Learning-based CNN Inference Offloading in Dynamic Edge Computing

TY - GEN

T1 - Graph Reinforcement Learning-based CNN Inference Offloading in Dynamic Edge Computing

AU - Li, Nan

AU - Iosifidis, Alexandros

AU - Zhang, Qi

PY - 2022

Y1 - 2022

N2 - This paper studies the computational offloading of CNN inference in dynamic multi-access edge computing (MEC) networks. To address the uncertainties in communication time and Edge servers' available capacity, we use early-exit mechanism to terminate the computation earlier to meet the deadline of inference tasks. We design a reward function to trade off the communication, computation and inference accuracy, and formu-late the offloading problem of CNN inference as a maximization problem with the goal of maximizing the average inference accuracy and throughput in long term. To solve the maxi-mization problem, we propose a graph reinforcement learning-based early-exit mechanism (GRLE), which outperforms the state-of-the-art work, deep reinforcement learning-based online offloading (DROO) and its enhanced method, DROO with early-exit mechanism (DROOE), under different dynamic scenarios. The experimental results show that G RLE achieves the average accuracy up to 3.41 x over graph reinforcement learning (GRL) and 1.45x over DROOE, which shows the advantages of GRLE for offloading decision-making in dynamic MEC.

AB - This paper studies the computational offloading of CNN inference in dynamic multi-access edge computing (MEC) networks. To address the uncertainties in communication time and Edge servers' available capacity, we use early-exit mechanism to terminate the computation earlier to meet the deadline of inference tasks. We design a reward function to trade off the communication, computation and inference accuracy, and formu-late the offloading problem of CNN inference as a maximization problem with the goal of maximizing the average inference accuracy and throughput in long term. To solve the maxi-mization problem, we propose a graph reinforcement learning-based early-exit mechanism (GRLE), which outperforms the state-of-the-art work, deep reinforcement learning-based online offloading (DROO) and its enhanced method, DROO with early-exit mechanism (DROOE), under different dynamic scenarios. The experimental results show that G RLE achieves the average accuracy up to 3.41 x over graph reinforcement learning (GRL) and 1.45x over DROOE, which shows the advantages of GRLE for offloading decision-making in dynamic MEC.

KW - CNN infer-ence

KW - Dynamic computation offloading

KW - Edge computing

KW - Graph reinforcement learning

KW - Service reliability

UR - http://www.scopus.com/inward/record.url?scp=85146920931&partnerID=8YFLogxK

U2 - 10.1109/GLOBECOM48099.2022.10001067

DO - 10.1109/GLOBECOM48099.2022.10001067

M3 - Article in proceedings

SP - 982

EP - 987

BT - 2022 IEEE Global Communications Conference, GLOBECOM 2022 - Proceedings

PB - IEEE

T2 - IEEE Global Communications Conference 2022

Y2 - 4 December 2022 through 8 December 2022

ER -