GPDS: A multi-agent deep reinforcement learning game for anti-jamming secure computing in MEC network

Chen, Miaojiang; Liu, Wei; Zhang, Ning; Li, Junling; Ren, Yingying; Yi, Meng; Liu, Anfeng

doi:10.1016/j.eswa.2022.118394

Cited by 27 publications

(9 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…But the DDPG is more difficult to train and sometimes unstable as it uses an actorcritic approach. Reinforcement learning was utilized in many topics, such as wireless (Chen et al [30]) and mobile edge computing networks (Chen et al [31,32]).…”

Section: Background and Reviewmentioning

confidence: 99%

A Deep Neural Network Algorithm for Linear-Quadratic Portfolio Optimization with MGARCH and Small Transaction Costs

Papanicolaou¹,

Hao²,

Krishnamurthy³

et al. 2023

Preprint

View full text Add to dashboard Cite

We analyze a fixed-point algorithm for reinforcement learning (RL) of optimal portfolio mean-variance preferences in the setting of multivariate generalized autoregressive conditional-heteroskedasticity (MGARCH) with a small penalty on trading. A numerical solution is obtained using a neural network (NN) architecture within a recursive RL loop. A fixed-point theorem proves that NN approximation error has a big-oh bound that we can reduce by increasing the number of NN parameters. The functional form of the trading penalty has a parameter > 0 that controls the magnitude of transaction costs. When is small, we can implement an NN algorithm based on the expansion of the solution in powers of . This expansion has a base term equal to a myopic solution with an explicit form, and a first-order correction term that we compute in the RL loop. Our expansion-based algorithm is stable, allows for fast computation, and outputs a solution that shows positive testing performance.

show abstract

Section: Background and Reviewmentioning

confidence: 99%

A Deep Neural Network Algorithm for Linear-Quadratic Portfolio Optimization with MGARCH and Small Transaction Costs

Papanicolaou¹,

Hao²,

Krishnamurthy³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In [29] the authors describe malicious interference countermeasures as a multi-user intelligent game and proposed Game with Post-Decision State (GPDS) anti-jamming secure computing using time-varying channels in the Mobile Edge Computing (MEC) networks. The Nash equilibrium gives the potential optimal channel selection strategy of a deep reinforcement learning multi-user random game with a proposed post-decision state.…”

Section: Literature Reviewmentioning

confidence: 99%

Received Power Based Unmanned Aerial Vehicles (UAVs) Jamming Detection and Nodes Classification Using Machine Learning

Aldosari¹

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

This paper presents a machine-learning method for detecting jamming UAVs and classifying nodes during jamming attacks on Wireless Sensor Networks (WSNs). Jamming is a type of Denial of Service (DoS) attack and intentional interference where a malicious node transmits a high-power signal to increase noise on the receiver side to disrupt the communication channel and reduce performance significantly. To defend and prevent such attacks, the first step is to detect them. The current detection approaches use centralized techniques to detect jamming, where each node collects information and forwards it to the base station. As a result, overhead and communication costs increased. In this work, we present a jamming attack and classify nodes into different categories based on their location to the jammer by employing a single node observer. As a result, we introduced a machine learning model that uses distance ratios and power received as features to detect such attacks. Furthermore, we considered several types of jammers transmitting at different power levels to evaluate the proposed metrics using MATLAB. With a detection accuracy of 99.7% for the k-nearest neighbors (KNN) algorithm and average testing accuracy of 99.9%, the presented solution is capable of efficiently and accurately detecting jamming attacks in wireless sensor networks.

show abstract

“…But the DDPG is more difficult to train and sometimes unstable as it uses an actor-critic approach. Reinforcement learning was utilized in many topics, such as wireless (Chen et al [30]) and mobile edge computing networks (Chen et al [31,32]).…”

Section: A Background and Reviewmentioning

confidence: 99%

A Deep Neural Network Algorithm for Linear-Quadratic Portfolio Optimization With MGARCH and Small Transaction Costs

Papanicolaou

Fu²,

Krishnamurthy³

et al. 2023

IEEE Access

View full text Add to dashboard Cite

We analyze a fixed-point algorithm for reinforcement learning (RL) of optimal portfolio mean-variance preferences in the setting of multivariate generalized autoregressive conditionalheteroskedasticity (MGARCH) with a small penalty on trading. A numerical solution is obtained using a neural network (NN) architecture within a recursive RL loop. A fixed-point theorem proves that NN approximation error has a big-oh bound that we can reduce by increasing the number of NN parameters. The functional form of the trading penalty has a parameter ϵ > 0 that controls the magnitude of transaction costs. When ϵ is small, we can implement an NN algorithm based on the expansion of the solution in powers of ϵ. This expansion has a base term equal to a myopic solution with an explicit form, and a firstorder correction term that we compute in the RL loop. Our expansion-based algorithm is stable, allows for fast computation, and outputs a solution that shows positive testing performance.

show abstract

GPDS: A multi-agent deep reinforcement learning game for anti-jamming secure computing in MEC network

Cited by 27 publications

References 47 publications

A Deep Neural Network Algorithm for Linear-Quadratic Portfolio Optimization with MGARCH and Small Transaction Costs

A Deep Neural Network Algorithm for Linear-Quadratic Portfolio Optimization with MGARCH and Small Transaction Costs

Received Power Based Unmanned Aerial Vehicles (UAVs) Jamming Detection and Nodes Classification Using Machine Learning

A Deep Neural Network Algorithm for Linear-Quadratic Portfolio Optimization With MGARCH and Small Transaction Costs

Contact Info

Product

Resources

About