Adversarial attacks on mobile malware detection

Maryam Shahpasand, Leonard Hamey, Dinusha Vatsalan, Minhui Xue

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionResearchpeer-review

Abstract

In recent years, machine learning approaches have been widely adopted for computer security tasks, including malware detection. Malware is a potent threat and an ongoing issue especially on smartphones which account for more than half of global web traffic. Although detection solutions are improving with the advances in machine learning techniques, they have been shown to be vulnerable to adversarial samples that carefully crafted perturbation enables them to evade detection. We propose a machine learning based model to attack malware classifiers leveraging the expressive capability of generative adversarial networks (GANs). We use GANs to generate effective adversarial samples by implying a threshold on the distortion amount on the generated samples. We show that the generated samples can bypass detection in 99% of attempts using a real Android application dataset.
LanguageEnglish
Title of host publicationAI4Mobile 2019 - 2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile
EditorsYang Liu, Lei Ma, Li Li, Minhui Xue
Place of PublicationPiscataway NJ
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages17-20
Number of pages4
ISBN (Electronic)9781728118116
ISBN (Print)9781728118123
DOIs
Publication statusPublished - 21 Mar 2019
Event2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile, AI4Mobile '19 - Hangzhou, China
Duration: 24 Feb 201924 Feb 2019

Conference

Conference2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile, AI4Mobile '19
CountryChina
CityHangzhou
Period24/02/1924/02/19

Fingerprint

Learning systems
Smartphones
Security of data
Classifiers
Malware

Keywords

  • Malware
  • Machine learning
  • Adversarial Machine Learning

Cite this

Shahpasand, M., Hamey, L., Vatsalan, D., & Xue, M. (2019). Adversarial attacks on mobile malware detection. In Y. Liu, L. Ma, L. Li, & M. Xue (Eds.), AI4Mobile 2019 - 2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile (pp. 17-20). [8672711] Piscataway NJ: Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/AI4Mobile.2019.8672711
Shahpasand, Maryam ; Hamey, Leonard ; Vatsalan, Dinusha ; Xue, Minhui. / Adversarial attacks on mobile malware detection. AI4Mobile 2019 - 2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile. editor / Yang Liu ; Lei Ma ; Li Li ; Minhui Xue. Piscataway NJ : Institute of Electrical and Electronics Engineers (IEEE), 2019. pp. 17-20
@inproceedings{b20477486bfe45c29aff2e8dba363e39,
title = "Adversarial attacks on mobile malware detection",
abstract = "In recent years, machine learning approaches have been widely adopted for computer security tasks, including malware detection. Malware is a potent threat and an ongoing issue especially on smartphones which account for more than half of global web traffic. Although detection solutions are improving with the advances in machine learning techniques, they have been shown to be vulnerable to adversarial samples that carefully crafted perturbation enables them to evade detection. We propose a machine learning based model to attack malware classifiers leveraging the expressive capability of generative adversarial networks (GANs). We use GANs to generate effective adversarial samples by implying a threshold on the distortion amount on the generated samples. We show that the generated samples can bypass detection in 99{\%} of attempts using a real Android application dataset.",
keywords = "Malware, Machine learning, Adversarial Machine Learning",
author = "Maryam Shahpasand and Leonard Hamey and Dinusha Vatsalan and Minhui Xue",
year = "2019",
month = "3",
day = "21",
doi = "10.1109/AI4Mobile.2019.8672711",
language = "English",
isbn = "9781728118123",
pages = "17--20",
editor = "Yang Liu and Lei Ma and Li Li and Minhui Xue",
booktitle = "AI4Mobile 2019 - 2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile",
publisher = "Institute of Electrical and Electronics Engineers (IEEE)",
address = "United States",

}

Shahpasand, M, Hamey, L, Vatsalan, D & Xue, M 2019, Adversarial attacks on mobile malware detection. in Y Liu, L Ma, L Li & M Xue (eds), AI4Mobile 2019 - 2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile., 8672711, Institute of Electrical and Electronics Engineers (IEEE), Piscataway NJ, pp. 17-20, 2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile, AI4Mobile '19, Hangzhou, China, 24/02/19. https://doi.org/10.1109/AI4Mobile.2019.8672711

Adversarial attacks on mobile malware detection. / Shahpasand, Maryam; Hamey, Leonard; Vatsalan, Dinusha; Xue, Minhui.

AI4Mobile 2019 - 2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile. ed. / Yang Liu; Lei Ma; Li Li; Minhui Xue. Piscataway NJ : Institute of Electrical and Electronics Engineers (IEEE), 2019. p. 17-20 8672711.

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionResearchpeer-review

TY - GEN

T1 - Adversarial attacks on mobile malware detection

AU - Shahpasand, Maryam

AU - Hamey, Leonard

AU - Vatsalan, Dinusha

AU - Xue, Minhui

PY - 2019/3/21

Y1 - 2019/3/21

N2 - In recent years, machine learning approaches have been widely adopted for computer security tasks, including malware detection. Malware is a potent threat and an ongoing issue especially on smartphones which account for more than half of global web traffic. Although detection solutions are improving with the advances in machine learning techniques, they have been shown to be vulnerable to adversarial samples that carefully crafted perturbation enables them to evade detection. We propose a machine learning based model to attack malware classifiers leveraging the expressive capability of generative adversarial networks (GANs). We use GANs to generate effective adversarial samples by implying a threshold on the distortion amount on the generated samples. We show that the generated samples can bypass detection in 99% of attempts using a real Android application dataset.

AB - In recent years, machine learning approaches have been widely adopted for computer security tasks, including malware detection. Malware is a potent threat and an ongoing issue especially on smartphones which account for more than half of global web traffic. Although detection solutions are improving with the advances in machine learning techniques, they have been shown to be vulnerable to adversarial samples that carefully crafted perturbation enables them to evade detection. We propose a machine learning based model to attack malware classifiers leveraging the expressive capability of generative adversarial networks (GANs). We use GANs to generate effective adversarial samples by implying a threshold on the distortion amount on the generated samples. We show that the generated samples can bypass detection in 99% of attempts using a real Android application dataset.

KW - Malware

KW - Machine learning

KW - Adversarial Machine Learning

UR - http://www.scopus.com/inward/record.url?scp=85064200375&partnerID=8YFLogxK

U2 - 10.1109/AI4Mobile.2019.8672711

DO - 10.1109/AI4Mobile.2019.8672711

M3 - Conference proceeding contribution

SN - 9781728118123

SP - 17

EP - 20

BT - AI4Mobile 2019 - 2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile

A2 - Liu, Yang

A2 - Ma, Lei

A2 - Li, Li

A2 - Xue, Minhui

PB - Institute of Electrical and Electronics Engineers (IEEE)

CY - Piscataway NJ

ER -

Shahpasand M, Hamey L, Vatsalan D, Xue M. Adversarial attacks on mobile malware detection. In Liu Y, Ma L, Li L, Xue M, editors, AI4Mobile 2019 - 2019 IEEE 1st International Workshop on Artificial Intelligence for Mobile. Piscataway NJ: Institute of Electrical and Electronics Engineers (IEEE). 2019. p. 17-20. 8672711 https://doi.org/10.1109/AI4Mobile.2019.8672711