Bit vector-coded simple CART structure for low latency traffic classification on FPGAs

Soylu, Tuncay; Erdem, Oğuzhan; Carus, Aydin

doi:10.1016/j.comnet.2019.106977

“…Switches or routers classify incoming packets to determine what action to take (e.g., forward, drop) . An active research area in networking is to propose classifiers with low inference latency (Chiu et al 2018;Soylu, Erdem, and Carus 2020;Rashelbach, Rottenstreich, and Silberstein 2020). We take a recently-proposed packet classifier for example, which was designed to optimize classification time and memory footprint.…”

Section: Methodsmentioning

confidence: 99%

“…Zinan Lin acknowledges the support of the Siemens FutureMakers Fellowship, the CMU Presidential Fellowship, and the Cylab Presidential Fellowship. This work used the Extreme Science and Engineering Discovery Environment (XSEDE) (Towns et al 2014), which is supported by National Science Foundation grant number ACI-1548562. Specifically, it used the Bridges system (Nystrom et al 2015), which is supported by NSF award number ACI-1445606, at the Pittsburgh Supercomputing Center (PSC).…”

Section: Acknowledgementsmentioning

confidence: 99%

RareGAN: Generating Samples for Rare Classes

Lin¹,

Liang²,

Fanti³

et al. 2022

Preprint

View full text Add to dashboard Cite

We study the problem of learning generative adversarial networks (GANs) for a rare class of an unlabeled dataset subject to a labeling budget. This problem is motivated from practical applications in domains including security (e.g., synthesizing packets for DNS amplification attacks), systems and networking (e.g., synthesizing workloads that trigger high resource usage), and machine learning (e.g., generating images from a rare class). Existing approaches are unsuitable, either requiring fully-labeled datasets or sacrificing the fidelity of the rare class for that of the common classes. We propose RareGAN, a novel synthesis of three key ideas: (1) extending conditional GANs to use labelled and unlabelled data for better generalization; (2) an active learning approach that requests the most useful labels; and (3) a weighted loss function to favor learning the rare class. We show that RareGAN achieves a better fidelity-diversity tradeoff on the rare class than prior work across different applications, budgets, rare class fractions, GAN losses, and architectures 1 .

show abstract

“…Switches or routers classify incoming packets to determine what action to take (e.g., forward, drop) (Liang et al 2019). An active research area in networking is to propose classifiers with low inference latency (Chiu et al 2018;Liang et al 2019;Soylu, Erdem, and Carus 2020;Rashelbach, Rottenstreich, and Silberstein 2020). We take a recently-proposed packet classifier (Liang et al 2019) for example, which was designed to optimize classification time and memory footprint.…”

Section: Methodsmentioning

confidence: 99%

RareGAN: Generating Samples for Rare Classes

Lin

¹

,

Hao

²

,

Fanti

³

et al. 2022

AAAI

View full text Add to dashboard Cite

We study the problem of learning generative adversarial networks (GANs) for a rare class of an unlabeled dataset subject to a labeling budget. This problem is motivated from practical applications in domains including security (e.g., synthesizing packets for DNS amplification attacks), systems and networking (e.g., synthesizing workloads that trigger high resource usage), and machine learning (e.g., generating images from a rare class). Existing approaches are unsuitable, either requiring fully-labeled datasets or sacrificing the fidelity of the rare class for that of the common classes. We propose RareGAN, a novel synthesis of three key ideas: (1) extending conditional GANs to use labelled and unlabelled data for better generalization; (2) an active learning approach that requests the most useful labels; and (3) a weighted loss function to favor learning the rare class. We show that RareGAN achieves a better fidelity-diversity tradeoff on the rare class than prior work across different applications, budgets, rare class fractions, GAN losses, and architectures.

show abstract

“…• CART [22] (Classification and Regression Trees) строит бинарное дерево решений с использованием критерия Джини. Деревья принятия решений и решения, построенные на их основе, являются одним из самых популярных способов решения задачи классификации трафика, так как среди их достоинств можно перечислить:…”

Section: дерево принятия решенийunclassified

A Survey of Network Traffic Classification

Getman¹,

Иконникова²

2020

Proceedings of ISP RAS

6

0

View full text Add to dashboard Cite

This survey is dedicated to the task of network traffic classification, particularly to the use of machine learning algorithms in this task. The survey begins with the description of the task, its variations and possible uses in real-world problems. It then proceeds to the description of the methods used historically to solve this task, their limitations and evolution of traffic making machine learning the main way to solve the problem. Then the most popular machine learning algorithms used in this task are described, with the examples of research papers, providing the insight into their advantages and disadvantages in relation to this field. The task of feature selection is discussed, followed by the more global problem of acquiring the suitable dataset to use in the research; some examples of such popular datasets and their descriptions are provided. The paper concludes with the outline of the current problems in this research area to be solved.

show abstract

Bit vector-coded simple CART structure for low latency traffic classification on FPGAs

Cited by 6 publications

References 9 publications

RareGAN: Generating Samples for Rare Classes

RareGAN: Generating Samples for Rare Classes

RareGAN: Generating Samples for Rare Classes

A Survey of Network Traffic Classification

Contact Info

Product

Resources

About