Which metric is commonly used to evaluate the adversarial robustness of a classification model?

The number of layers in the neural network.

The time taken to train the model.

The size of the training dataset.

Accuracy on a set of adversarial examples.

A team is testing the adversarial robustness of a model for autonomous vehicle sign recognition. They have a limited budget and need to evaluate the model's vulnerability to both white-box and black-box attacks. Which testing approach is most cost-effective and comprehensive?

Perform black-box attacks by querying the model through an API

Only use adversarial training to improve robustness without testing

Use random noise to test the model's sensitivity

Perform white-box attacks using FGSM and PGD on a subset of the test set

A company uses an AI model for face recognition in security systems. They want to defend against physical adversarial attacks, such as wearing specially crafted glasses to fool the system. Which defense mechanism is most appropriate?

Input preprocessing (e.g., image cropping)

Rate limiting access to the system

Adversarial training with physical-world perturbations

Certified Tester AI Testing

Domain 9Objective 1

Adversarial Attacks CT-AI Practice Questions (Page 1)

Part of the Domain 9: Methods and Techniques for the Testing of AI-Based Systems domain, which makes up ~13% of our current practice bank. ISTQB does not publish an official question count, but from its 60-minute exam (~25–40 total, ~3–5 in this domain), expect 1–1 from this objective — we provide 24 practice questions to prepare you well beyond it. (estimate)

24questions here

5free pages

7concepts

Questions 1–5

1foundation · easy
Which type of adversarial attack involves an attacker providing malicious data during the training phase to corrupt the model's learning?
Select an answer first
2application · medium
A data scientist is testing the robustness of a model for image classification. They want to generate adversarial examples that are as close to the original images as possible while still causing misclassification. Which method is most suitable?
Select an answer first
3foundation · easy
Which metric is commonly used to evaluate the adversarial robustness of a classification model?
Select an answer first
4expert · hard
A team is testing the adversarial robustness of a model for autonomous vehicle sign recognition. They have a limited budget and need to evaluate the model's vulnerability to both white-box and black-box attacks. Which testing approach is most cost-effective and comprehensive?
Select an answer first
5application · medium
A company uses an AI model for face recognition in security systems. They want to defend against physical adversarial attacks, such as wearing specially crafted glasses to fool the system. Which defense mechanism is most appropriate?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to CT-AI

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by ISTQB. “CT-AI” is a trademark of its owner, used for identification only.

Adversarial Attacks CT-AI Practice Questions (Page 1)

Questions 1–5

Which type of adversarial attack involves an attacker providing malicious data during the training phase to corrupt the model's learning?

A data scientist is testing the robustness of a model for image classification. They want to generate adversarial examples that are as close to the original images as possible while still causing misclassification. Which method is most suitable?

Which metric is commonly used to evaluate the adversarial robustness of a classification model?

A team is testing the adversarial robustness of a model for autonomous vehicle sign recognition. They have a limited budget and need to evaluate the model's vulnerability to both white-box and black-box attacks. Which testing approach is most cost-effective and comprehensive?

A company uses an AI model for face recognition in security systems. They want to defend against physical adversarial attacks, such as wearing specially crafted glasses to fool the system. Which defense mechanism is most appropriate?

Keep studying — Domain 9: Methods and Techniques for the Testing of AI-Based Systems