Generating Adversarial Phishing Websites to Evade Machine Learning-based Anti-Phishing Detectors: A Reinforcement Learning Approach

Yang Gao; Sagar Samtani; Hongyi Zhu; Benjamin M. Ampel; Yidong Chai

Generating Adversarial Phishing Websites to Evade Machine Learning-based Anti-Phishing Detectors: A Reinforcement Learning Approach

Yang Gao, Sagar Samtani, Hongyi Zhu, Benjamin M. Ampel, Yidong Chai

Last updated on Jun 9, 2026

In Plain Terms

This workshop paper studies how attackers could automatically create phishing websites designed to slip past machine learning detectors. Using a reinforcement learning approach, the authors generate adversarial phishing pages to expose weak spots in current anti-phishing defenses and motivate the design of more robust detection systems.

Artifacts

No artifacts listed yet.

Related Papers

Cite this

APA Gao, Y., Samtani, S., Zhu, H., Ampel, B. M., & Chai, Y. (2023). Generating Adversarial Phishing Websites to Evade Machine Learning-based Anti-Phishing Detectors: A Reinforcement Learning Approach. In INFORMS Workshop on Data Science (WDS).

BibTeX

@inproceedings{gao2023generating,
  title={Generating Adversarial Phishing Websites to Evade Machine Learning-based Anti-Phishing Detectors: A Reinforcement Learning Approach},
  author={Yang Gao and Sagar Samtani and Hongyi Zhu and Benjamin M. Ampel and Yidong Chai},
  booktitle={INFORMS Workshop on Data Science (WDS)},
  year={2023}
}

Phishing Reinforcement Learning Adversarial Machine Learning Cybersecurity

Generating Adversarial Phishing Websites to Evade Machine Learning-based Anti-Phishing Detectors: A Reinforcement Learning Approach

Artifacts

Related Papers

Cite this

Benjamin M. Ampel

Assistant Professor in Computer Information Systems and Director, Center for CyberAI Research (CCAIR)