Generative Adversarial Networks (GANs) in networking: A comprehensive survey &amp; evaluation

Navidan, Hojjat; Moshiri, Parisa Fard; Nabati, Mohammad; Shahbazian, Reza; Ghorashi, Seyed Ali; Shah-Mansouri, Vahid; Windridge, David

doi:10.1016/j.comnet.2021.108149

Cited by 66 publications

(19 citation statements)

References 101 publications

(106 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A straightforward semi-supervised approach can be present by combining the supervised loss and unsupervised loss throughout training, which has been demonstrated useful for many tasks i.e., classification, detection, etc. This observation shows that a straightforward and effective semisupervised approach could be supplied by merging a supervised cost function and an unsupervised GAN function [14].…”

Section: Semi-supervised Generative Approachesmentioning

confidence: 84%

Semi-supervised Deep Learning for Secure Internet of Things

Abdel‐Basset

Moustafa

Hawash

et al. 2021

Deep Learning Techniques for IoT Security and Privacy

View full text Add to dashboard Cite

Previous chapters demonstrate the great success achieved by principally in supervised settings, by leveraging a larger volume of precisely annotated dataset. Nevertheless, annotated data instances are frequently complicated, costly, or laborious to acquire. The annotation procedure typically necessitates many experts' efforts, which is the foremost drawback for training a brilliant supervised deep model. when a small number of annotated observations are obtainable, it could be interesting to develop a thriving learning solution. On the opposite side, the unlabeled data is generally plentiful and could be simply or cheaply attained. Therefore, it is advantageous to use a huge number of unannotated data for enhancing the learning performance given a little amount of annotated data. motivated by that, semi-supervised learning is regarded as an interesting research field concerned with developing efficient models trained with both labeled and unlabeled data. it also seeks to improve the learning performance by leveraging extra unlabeled examples contrasted to supervised approaches based merely on the annotated samples. semi-supervised solutions could easily obtain by extending supervised or unsupervised learning algorithms. Therefore, this chapter emphasis discussing the potential of a state-of-the-art deep learning solution that can be exploited for securing the Internet of Things (IoT) by leveraging partially labeled data.This chapter covers the following topics.

show abstract

Section: Semi-supervised Generative Approachesmentioning

confidence: 84%

Semi-supervised Deep Learning for Secure Internet of Things

Abdel‐Basset

Moustafa

Hawash

et al. 2021

Deep Learning Techniques for IoT Security and Privacy

View full text Add to dashboard Cite

show abstract

“…Over the past, a few different research works have targeted the generation of synthetic network traffic using GANs 18 , although the majority of them only propose data augmentation solutions that are not applicable in scenarios in which data privacy must be guaranteed as they use a combination of real and synthetic data.…”

Section: Related Workmentioning

confidence: 99%

Synthetic flow-based cryptomining attack generation through Generative Adversarial Networks

Mozó

González-Prieto

Pastor

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Due to the growing rise of cyber attacks in the Internet, the demand of accurate intrusion detection systems (IDS) to prevent these vulnerabilities is increasing. To this aim, Machine Learning (ML) components have been proposed as an efficient and effective solution. However, its applicability scope is limited by two important issues: (i) the shortage of network traffic data datasets for attack analysis, and (ii) the data privacy constraints of the data to be used. To overcome these problems, Generative Adversarial Networks (GANs) have been proposed for synthetic flow-based network traffic generation. However, due to the ill-convergence of the GAN training, none of the existing solutions can generate high-quality fully synthetic data that can totally substitute real data in the training of ML components. In contrast, they mix real with synthetic data, which acts only as data augmentation components, leading to privacy breaches as real data is used. In sharp contrast, in this work we propose a novel and deterministic way to measure the quality of the synthetic data produced by a GAN both with respect to the real data and to its performance when used for ML tasks. As a by-product, we present a heuristic that uses these metrics for selecting the best performing generator during GAN training, leading to a novel stopping criterion, which can be applied even when different types of synthetic data are to be used in the same ML task. We demonstrate the adequacy of our proposal by generating synthetic cryptomining attacks and normal traffic flow-based data using an enhanced version of a Wasserstein GAN. The results evidence that the generated synthetic network traffic can completely replace real data when training a ML-based cryptomining detector, obtaining similar performance and avoiding privacy violations, since real data is not used in the training of the ML-based detector.

show abstract

“…The research pertaining to GANs can be classified into two broad groups. The first group is concerned with the application of GANs to different real-world problems in different domains like computer vision [19], natural language processing [20], biology [21], astronomy [22], networking [23], and other areas. GANs have been successful in generating betterresolution samples from poor-resolution ones [24], generating images and videos from textual descriptions [25,26], and Image-to-Image Translation [27].…”

Section: Related Workmentioning

confidence: 99%

Mathematical analysis of loss function of GAN and its loss function variants

2022

IJATEE

View full text Add to dashboard Cite

Generative adversarial networks (GANs) have turned up as the most widely used approaches for creating realistic samples. They're the effective latent variable models for learning complex real distributions. However, despite their enormous success and popularity, the process of training GANs remains challenging and suffers from a number of failures. These failures include mode collapse where the generator generates the same set of output for different inputs which finally leads to loss of diversity; non-convergence because of the diverging and oscillatory behaviors of both generator and discriminator while being trained; and vanishing or exploding gradients due to which either no learning or extremely slow learning takes place. In the past years, a variety of strategies for stabilizing GAN training have been explored which includes modified architectures, loss functions, and other methods. The choice of loss function has been found to be the most crucial part of the GAN model because it influences the vanishing gradient and model collapse directly. Viewing these loss functions as divergence minimization has provided a rich avenue of development. All of these factors make GAN training inherently unstable, and this instability is difficult to comprehend mathematically. This paper intends to provide a thorough mathematical explanation of these divergence minimization functions. It illustrates in great detail the two variants of the loss functions of the original GAN, their optimization to kullback-leibler (KL) divergence and jensen-shannon (JS) divergence along with their shortcomings. It also describes the loss functions of the different loss function GAN variants that have been proposed to mitigate these shortcomings as well as their minimization. The original GAN and its loss function variants have also been implemented using the standard MNIST, Fashion-MNIST, and CIFAR-10 datasets.

show abstract

Generative Adversarial Networks (GANs) in networking: A comprehensive survey & evaluation

Cited by 66 publications

References 101 publications

Semi-supervised Deep Learning for Secure Internet of Things

Semi-supervised Deep Learning for Secure Internet of Things

Synthetic flow-based cryptomining attack generation through Generative Adversarial Networks

Mathematical analysis of loss function of GAN and its loss function variants

Contact Info

Product

Resources

About