Learning for Computation Offloading in Mobile Edge Computing

Dinh, Thinh Quang; Lã, Quang Duy; Quek, Tony Q. S.; Shin, Hyundong

doi:10.1109/tcomm.2018.2866572

Cited by 184 publications

(79 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This process continues until a specific stopping criterion is met. RL has been applied to address issues in CC and EC [126]- [131]. • Deep RL (DRL) can be seen as a class of new efficient learning algorithm by combining DL with RL [132]- [134].…”

Section: Lbsmentioning

confidence: 99%

“…Experimental results show that the proposed algorithm outperforms cloudlet-based dynamic task offloading in terms of energy consumption and completion time. Dinh et al [131] proposed a RL-based computation offloading scheme in MEC to reduce energy consumption. They studied multi-user multi-edge-node computation offloading problem, and formulated it as a non-cooperative game where each user maximizes its own utility.…”

Section: Lbs In CC and Ecmentioning

confidence: 99%

See 1 more Smart Citation

A Review on Computational Intelligence Techniques in Cloud and Edge Computing

Asim

Wang

et al. 2020

IEEE Trans. Emerg. Top. Comput. Intell.

View full text Add to dashboard Cite

Cloud computing (CC) is a centralized computing paradigm that accumulates resources centrally and provides these resources to users through Internet. Although CC holds a large number of resources, it may not be acceptable by real-time mobile applications, as it is usually far away from users geographically. On the other hand, edge computing (EC), which distributes resources to the network edge, enjoys increasing popularity in the applications with low-latency and high-reliability requirements. EC provides resources in a decentralized manner, which can respond to users' requirements faster than the normal CC, but with limited computing capacities. As both CC and EC are resource-sensitive, several big issues arise, such as how to conduct job scheduling, resource allocation, and task offloading, which significantly influence the performance of the whole system. To tackle these issues, many optimization problems have been formulated. These optimization problems usually have complex properties, such as non-convexity and NP-hardness, which may not be addressed by the traditional convex optimization-based solutions. Computational intelligence (CI), consisting of a set of nature-inspired computational approaches, recently exhibits great potential in addressing these optimization problems in CC and EC. This paper provides an overview of research problems in CC and EC and recent progresses in addressing them with the help of CI techniques. Informative discussions and future research trends are also presented, with the aim of offering insights to the readers and motivating new research directions.

show abstract

Section: Lbsmentioning

confidence: 99%

Section: Lbs In CC and Ecmentioning

confidence: 99%

A Review on Computational Intelligence Techniques in Cloud and Edge Computing

Asim

Wang

et al. 2020

IEEE Trans. Emerg. Top. Comput. Intell.

View full text Add to dashboard Cite

show abstract

“…Sardellitti et al [14] considered the joint optimization of radio and computational resources for computation offloading in a dense deployment scenario, in the presence of intercell interference. In order to reduce the signalling overhead encountered in FO MEC systems, decentralized algorithms [15] have also been proposed. Wang et al [16] conceived a decentralized algorithm based on the alternating direction multiplier method (ADMM) for computation offloading, resource allocation and Internet content caching optimization in heterogeneous MEC-aided wireless networks.…”

Section: A Prior Workmentioning

confidence: 99%

Energy-Efficient Resource Allocation for Latency-Sensitive Mobile Edge Computing

Chen

Zhao

et al. 2020

IEEE Trans. Veh. Technol.

View full text Add to dashboard Cite

Resource allocation algorithms are conceived for minimizing the energy consumption of multiuser mobile edge computing (MEC) systems operating in the face of interference channels, and where mobile users can offload their latencysensitive tasks to the mobile edge server via a base station (BS). Latency-sensitive applications that benefit from MEC services can be divided into two major classes: 1) applications requiring uninterrupted execution and that cannot be fragmented and therefore require full offloading (FO); 2) applications which can benefit from fractional or partial offloading (PO). For each class of applications, we first formulate a joint optimization problem where the aim is to minimize the overall energy consumption across the sub-network subject to latency, transmission quality, computational budget and transmit power constraints. The proposed optimization problems are nonconvex, tightly coupled, and consequently challenging to solve. By exploiting binary relaxation, smooth approximation and auxiliary variables, we convert these problems into more tractable forms and subsequently develop novel algorithms based on the concave-convex procedure (CCCP) to solve them. Furthermore, by incorporating a measure of user priority, a reduced-complexity solution is proposed for the FO scheme. The benefits of our energy-efficient resource allocation algorithms for latency-sensitive MEC are demonstrated through simulations.

show abstract

“…Markov decision process (MDP) can be also applied to the analysis and design of dynamic control of computation offloading [19]. Furthermore, [20] demonstrated how dynamic computation offloading policies can be learned by reinforcement learning (RL)-based algorithm with no prior knowledge of the system.…”

Section: Introductionmentioning

confidence: 99%

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

Zhao

Wang

2020

J Wireless Com Network

135

View full text Add to dashboard Cite

Mobile edge computing (MEC) emerges recently as a promising solution to relieve resource-limited mobile devices from computation-intensive tasks, which enables devices to offload workloads to nearby MEC servers and improve the quality of computation experience. In this paper, an MEC enabled multi-user multi-input multi-output (MIMO) system with stochastic wireless channels and task arrivals is considered. In order to minimize long-term average computation cost in terms of power consumption and buffering delay at each user, a deep reinforcement learning (DRL)-based dynamic computation offloading strategy is investigated to build a scalable system with limited feedback. Specifically, a continuous action space-based DRL approach named deep deterministic policy gradient (DDPG) is adopted to learn decentralized computation offloading policies at all users respectively, where local execution and task offloading powers will be adaptively allocated according to each user’s local observation. Numerical results demonstrate that the proposed DDPG-based strategy can help each user learn an efficient dynamic offloading policy and also verify the superiority of its continuous power allocation capability to policies learned by conventional discrete action space-based reinforcement learning approaches like deep Q-network (DQN) as well as some other greedy strategies with reduced computation cost. Besides, power-delay tradeoff for computation offloading is also analyzed for both the DDPG-based and DQN-based strategies.

show abstract

Learning for Computation Offloading in Mobile Edge Computing

Cited by 184 publications

References 33 publications

A Review on Computational Intelligence Techniques in Cloud and Edge Computing

A Review on Computational Intelligence Techniques in Cloud and Edge Computing

Energy-Efficient Resource Allocation for Latency-Sensitive Mobile Edge Computing

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

Contact Info

Product

Resources

About