Which of the following is an example of input preprocessing as a defense against adversarial attacks?

Adding more layers to the neural network to increase its capacity.

Changing the model's activation function to a different type.

Smoothing the input image to reduce the effect of small perturbations.

Using a larger training dataset to improve accuracy.

A data scientist is testing the robustness of a deep learning model for malware detection. They want to generate adversarial examples that are minimally perturbed but still cause misclassification. Which method is most appropriate for this purpose?

Data augmentation with rotations

Projected Gradient Descent (PGD)

Fast Gradient Sign Method (FGSM)

What is the core idea behind the Fast Gradient Sign Method (FGSM) for generating adversarial examples?

It iteratively refines the perturbation by taking multiple small steps in the gradient direction.

It adds a small perturbation in the direction of the gradient of the loss function with respect to the input.

It replaces the input with a completely different image that is classified as the target class.

It uses a random search to find inputs that cause misclassification.

In the context of adversarial attacks, what does a threat model primarily describe?

The defense mechanisms that are implemented to protect the AI system.

The performance metrics used to evaluate the AI system's accuracy.

The attacker's capabilities, goals, and the assumptions about the AI system's environment.

The specific algorithm used to generate adversarial examples.

Certified Tester AI Testing

Domain 9Objective 1

Adversarial Attacks CT-AI Practice Questions (Page 3)

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 11–15

11foundation · easy
Which of the following is an example of input preprocessing as a defense against adversarial attacks?
Select an answer first
12application · medium
A data scientist is testing the robustness of a deep learning model for malware detection. They want to generate adversarial examples that are minimally perturbed but still cause misclassification. Which method is most appropriate for this purpose?
Select an answer first
13foundation · easy
An attacker creates a slightly altered image of a stop sign that the AI system misclassifies as a speed limit sign. This is an example of which type of adversarial attack?
Select an answer first
14foundation · easy
What is the core idea behind the Fast Gradient Sign Method (FGSM) for generating adversarial examples?
Select an answer first
15foundation · easy
In the context of adversarial attacks, what does a threat model primarily describe?
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.