MAB-Malware: A Reinforcement Learning Framework for Attacking Static Malware Classifiers Article Swipe

MAKE IMPACT

Wei Song , Xuezixiang Li , Sadia Afroz , Deepali Garg , Dmitry Kuznetsov , Heng Yin · YOU? · · 2020 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2003.03100

Modern commercial antivirus systems increasingly rely on machine learning to keep up with the rampant inflation of new malware. However, it is well-known that machine learning models are vulnerable to adversarial examples (AEs). Previous works have shown that ML malware classifiers are fragile to the white-box adversarial attacks. However, ML models used in commercial antivirus products are usually not available to attackers and only return hard classification labels. Therefore, it is more practical to evaluate the robustness of ML models and real-world AVs in a pure black-box manner. We propose a black-box Reinforcement Learning (RL) based framework to generate AEs for PE malware classifiers and AV engines. It regards the adversarial attack problem as a multi-armed bandit problem, which finds an optimal balance between exploiting the successful patterns and exploring more varieties. Compared to other frameworks, our improvements lie in three points. 1) Limiting the exploration space by modeling the generation process as a stateless process to avoid combination explosions. 2) Due to the critical role of payload in AE generation, we design to reuse the successful payload in modeling. 3) Minimizing the changes on AE samples to correctly assign the rewards in RL learning. It also helps identify the root cause of evasions. As a result, our framework has much higher black-box evasion rates than other off-the-shelf frameworks. Results show it has over 74\%--97\% evasion rate for two state-of-the-art ML detectors and over 32\%--48\% evasion rate for commercial AVs in a pure black-box setting. We also demonstrate that the transferability of adversarial attacks among ML-based classifiers is higher than the attack transferability between purely ML-based and commercial AVs.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2003.03100

PDF

https://arxiv.org/pdf/2003.03100

OA Status

green

Cited By

22

References

53

Related Works

10

OpenAlex ID

https://openalex.org/W3158179156

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W3158179156

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2003.03100

Digital Object Identifier

Title

MAB-Malware: A Reinforcement Learning Framework for Attacking Static Malware Classifiers

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2020

Year of publication

Publication date

2020-03-06

Full publication date if available

Authors

Wei Song, Xuezixiang Li, Sadia Afroz, Deepali Garg, Dmitry Kuznetsov, Heng Yin

List of authors in order

Landing page

https://arxiv.org/abs/2003.03100

Publisher landing page

PDF URL

https://arxiv.org/pdf/2003.03100

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2003.03100

Direct OA link when available

Concepts

Malware, Evasion (ethics), Computer science, Reinforcement learning, Machine learning, Artificial intelligence, Payload (computing), Adversarial system, Black box, System call, Computer security, Operating system, Immune system, Immunology, Network packet, Biology

Top concepts (fields/topics) attached by OpenAlex

Cited by

22

Total citation count in OpenAlex

Citations by year (recent)

2025: 2, 2024: 5, 2023: 6, 2022: 5, 2021: 4

Per-year citation counts (last 5 years)

References (count)

53

Number of works referenced by this work

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload