LyDROO
Lyapunovguided Deep Reinforcement Learning for Stable Online Computation Offloading in MobileEdge Computing Networks
PyTorch code to reproduce LyDROO algorithm [1], which is an online computation offloading algorithm to maximize the network data processing capability subject to the longterm data queue stability and average power constraints. It applies Lyapunov optimization to decouple the multistage stochastic MINLP into deterministic perframe MINLP subproblems and solves each subproblem via DROO algorithm. It includes:

memory.py: the codes of Deep Reinforcement Learning

ResourceAllocation: Algorithms for resource allocation

LyDROO.py: run this file for LyDROO
About our works
 Suzhi Bi, Liang Huang, and Yingjun Angela Zhang, ``Lyapunovguided Deep Reinforcement Learning for Stable Online Computation Offloading in MobileEdge Computing Networks'', IEEE Transactions on Wireless Communications, 2021, doi:10.1109/TWC.2021.3085319.
About authors

Suzhi BI, bsz AT szu.edu.cn

Liang HUANG, lianghuang AT zjut.edu.cn

Ying Jun (Angela) Zhang, yjzhang AT ie.cuhk.edu.hk
How the code works
 For LyDROO algorithm, run the file, LyDROO.py