GRAPHSEC – Advancing the Application of AI/ML to Network Security through Graph Neural Networks

Pedro Casas-Hernandez; Juan Vanerio; Johanna  Ullrich; Mislav  Findrik; Pere Barlet-Ros

doi:10.1007/978-3-031-36183-8_5

GRAPHSEC – Advancing the Application of AI/ML to Network Security through Graph Neural Networks

Pedro Casas-Hernandez (Author and Speaker), Juan Vanerio, Johanna Ullrich, Mislav Findrik, Pere Barlet-Ros

Data Science & Artifical Intelligence

Research output: Chapter in Book or Conference Proceedings › Conference Proceedings with Oral Presentation › peer-review

Abstract

The application of Artificial Intelligence (AI) and Machine Learning (ML) to network security (AI4SEC) is paramount against cybercrime. While AI/ML is today mainstream in domains such as computer vision and speech recognition, it has produced below-par results in AI4SEC. Solutions do not properly generalize, are ineffective in real deployments, and are vulnerable to adversarial attacks. A fundamental limitation is the lack of AI/ML technology specific to network security. Network security data is intrinsically relational, and graph-structured data representations and Graph Neural Networks (GNNs) have the potential to drastically advance the AI4SEC domain. In this positioning paper we propose GRAPHSEC, a research agenda to systematically integrate GNNs in AI4SEC. We structure the state of the art in AI4SEC and on the application of GNNs to network security applications, elaborate on the benefits and challenges faced by GRAPHSEC, and propose a research agenda to advance the AI4SEC domain through GNNs.

Original language	English
Title of host publication	MLN 2022: Machine Learning for Networking.
Editors	Éric Renault, Paul Mühlethaler
Pages	56-71
Volume	13767
ISBN (Electronic)	978-3-031-36183-8
DOIs	https://doi.org/10.1007/978-3-031-36183-8_5
Publication status	Published - 7 Jul 2023
Event	5th International Conference on Machine Learning for Networking, MLN 2022 - Duration: 28 Nov 2022 → 30 Nov 2022

Conference

Conference	5th International Conference on Machine Learning for Networking, MLN 2022
Period	28/11/22 → 30/11/22

Research Field

Former Research Field - Data Science

Keywords

Network Security
AI4SEC
Graph Neural Networks

Access to Document

10.1007/978-3-031-36183-8_5

Cite this

@inproceedings{5b19e750561a4191874632d9025bbaab,

title = "GRAPHSEC – Advancing the Application of AI/ML to Network Security through Graph Neural Networks",

abstract = "The application of Artificial Intelligence (AI) and Machine Learning (ML) to network security (AI4SEC) is paramount against cybercrime. While AI/ML is today mainstream in domains such as computer vision and speech recognition, it has produced below-par results in AI4SEC. Solutions do not properly generalize, are ineffective in real deployments, and are vulnerable to adversarial attacks. A fundamental limitation is the lack of AI/ML technology specific to network security. Network security data is intrinsically relational, and graph-structured data representations and Graph Neural Networks (GNNs) have the potential to drastically advance the AI4SEC domain. In this positioning paper we propose GRAPHSEC, a research agenda to systematically integrate GNNs in AI4SEC. We structure the state of the art in AI4SEC and on the application of GNNs to network security applications, elaborate on the benefits and challenges faced by GRAPHSEC, and propose a research agenda to advance the AI4SEC domain through GNNs.",

keywords = "Network Security, AI4SEC, Graph Neural Networks",

author = "Pedro Casas-Hernandez and Juan Vanerio and Johanna Ullrich and Mislav Findrik and Pere Barlet-Ros",

year = "2023",

month = jul,

day = "7",

doi = "10.1007/978-3-031-36183-8_5",

language = "English",

isbn = "978-3-031-36182-1",

volume = "13767",

pages = "56--71",

editor = "{\'E}ric Renault and Paul M{\"u}hlethaler",

booktitle = "MLN 2022: Machine Learning for Networking.",

note = "5th International Conference on Machine Learning for Networking, MLN 2022 ; Conference date: 28-11-2022 Through 30-11-2022",

}

Casas-Hernandez, P, Vanerio, J, Ullrich, J, Findrik, M & Barlet-Ros, P 2023, GRAPHSEC – Advancing the Application of AI/ML to Network Security through Graph Neural Networks. in É Renault & P Mühlethaler (eds), MLN 2022: Machine Learning for Networking.. vol. 13767, pp. 56-71, 5th International Conference on Machine Learning for Networking, MLN 2022, 28/11/22. https://doi.org/10.1007/978-3-031-36183-8_5

GRAPHSEC – Advancing the Application of AI/ML to Network Security through Graph Neural Networks. / Casas-Hernandez, Pedro (Author and Speaker); Vanerio, Juan; Ullrich, Johanna et al.
MLN 2022: Machine Learning for Networking.. ed. / Éric Renault; Paul Mühlethaler. Vol. 13767 2023. p. 56-71.

Research output: Chapter in Book or Conference Proceedings › Conference Proceedings with Oral Presentation › peer-review

TY - GEN

T1 - GRAPHSEC – Advancing the Application of AI/ML to Network Security through Graph Neural Networks

AU - Vanerio, Juan

AU - Ullrich, Johanna

AU - Findrik, Mislav

AU - Barlet-Ros, Pere

A2 - Casas-Hernandez, Pedro

A2 - Renault, Éric

A2 - Mühlethaler, Paul

PY - 2023/7/7

Y1 - 2023/7/7

N2 - The application of Artificial Intelligence (AI) and Machine Learning (ML) to network security (AI4SEC) is paramount against cybercrime. While AI/ML is today mainstream in domains such as computer vision and speech recognition, it has produced below-par results in AI4SEC. Solutions do not properly generalize, are ineffective in real deployments, and are vulnerable to adversarial attacks. A fundamental limitation is the lack of AI/ML technology specific to network security. Network security data is intrinsically relational, and graph-structured data representations and Graph Neural Networks (GNNs) have the potential to drastically advance the AI4SEC domain. In this positioning paper we propose GRAPHSEC, a research agenda to systematically integrate GNNs in AI4SEC. We structure the state of the art in AI4SEC and on the application of GNNs to network security applications, elaborate on the benefits and challenges faced by GRAPHSEC, and propose a research agenda to advance the AI4SEC domain through GNNs.

AB - The application of Artificial Intelligence (AI) and Machine Learning (ML) to network security (AI4SEC) is paramount against cybercrime. While AI/ML is today mainstream in domains such as computer vision and speech recognition, it has produced below-par results in AI4SEC. Solutions do not properly generalize, are ineffective in real deployments, and are vulnerable to adversarial attacks. A fundamental limitation is the lack of AI/ML technology specific to network security. Network security data is intrinsically relational, and graph-structured data representations and Graph Neural Networks (GNNs) have the potential to drastically advance the AI4SEC domain. In this positioning paper we propose GRAPHSEC, a research agenda to systematically integrate GNNs in AI4SEC. We structure the state of the art in AI4SEC and on the application of GNNs to network security applications, elaborate on the benefits and challenges faced by GRAPHSEC, and propose a research agenda to advance the AI4SEC domain through GNNs.

KW - Network Security

KW - AI4SEC

KW - Graph Neural Networks

U2 - 10.1007/978-3-031-36183-8_5

DO - 10.1007/978-3-031-36183-8_5

M3 - Conference Proceedings with Oral Presentation

SN - 978-3-031-36182-1

VL - 13767

SP - 56

EP - 71

BT - MLN 2022: Machine Learning for Networking.

T2 - 5th International Conference on Machine Learning for Networking, MLN 2022

Y2 - 28 November 2022 through 30 November 2022

ER -

GRAPHSEC – Advancing the Application of AI/ML to Network Security through Graph Neural Networks

Abstract

Conference

Research Field

Keywords

Access to Document

Fingerprint

Cite this