[FEEDBACK]: Description of ML04 Membership Inference Attack

[dferri]

Type

General Feedback

What would you like to report?

https://github.com/OWASP/www-project-machine-learning-security-top-10/blob/f1cf662ca9ce5cfcd4c72ab8d4bff91ea64f46d7/docs/ML04_2023-Membership_Inference_Attack.md?plain=1#L27

Here, the documentation states that an attacker manipulates the model’s training data, but from my understanding the objective of a membership inference attack is to "[...] predict whether or not a particular example was contained in the model’s training dataset.", so the attacker shouldn't have access to the training data.

I can create a pull request to update the documentation. Let me know if you'd like me to proceed.

Code of Conduct

• [X] I agree to follow this project's Code of Conduct