Question 1

An insurance company uses two different AI models for policy approval, one for Group A and another for Group B. The confusion matrices for each group are as follows:

Group A Confusion Matrix:

True Positives: 50

False Positives: 10

False Negatives: 40

True Negatives: 90

Group B Confusion Matrix:

True Positives: 30

False Positives: 20

False Negatives: 70

True Negatives: 80

What fairness technique should the company use to balance the performance between the two groups?

Options :

A : Equalize the number of true positives for both groups.

B : Adjust thresholds to achieve similar false positive rates for both groups.

C : Apply a higher penalty for false negatives in Group A.

D : Implement demographic parity by ensuring equal acceptance rates for both groups.

Answer: D

Question 2

A problem-solving AI model is being developed to solve the Tower of Hanoi puzzle by systematically planning moves to achieve the final goal. The model needs to evaluate intermediate steps to clear disks strategically for future moves. Which approach would best help the model to complete this task?

Options :

A : Tactical lookahead, as it considers each immediate move without long-term planning.

B : Strategic planning, by identifying intermediate moves that contribute to the end goal.

C : Reinforcement learning, by rewarding moves that lead to progress in the puzzle.

D : Randomized search, by exploring different moves without a predefined plan.

Answer: B

Question 3

A financial institution finds that the system surrounding its risk model no longer aligns with its IT infrastructure requirements. To align with new standards, which adaptation task should they focus on?

Options :

A : Core adaptation (Ct)

B : Model adaptation (Mt)

C : Process review

D : System adaptation (St)

Answer: D

Question 4

A government agency uses an AI system to determine eligibility for welfare benefits but faces public concerns about fairness due to its black-box nature. What is the best step to increase transparency?

Options :

A : Provide simplified output without explanation

B : Implement model-agnostic interpretability techniques

C :

Replace the system with manual processing

D : Use the system without public disclosure

Answer: B

Question 5

An insurance company is using a neural network for classifying claims as "Fraudulent" or "Non-Fraudulent." They decide to use ReLU (Rectified Linear Unit) as the activation function for the hidden layers. What is the primary purpose of using an activation function like ReLU in this neural network?

Options :

A : To scale the output between 0 and 1

B : To introduce non-linearity

C : To calculate weights in the hidden layers

D : To set bias terms to zero

Answer: B

Confidently Practice Online with Free RAI Exam Cram

Buying Options: