Machine Learning Ethics: Fairness and Discrimination

Question 1

In machine learning, informed consent involves:

Accepted Answer

Giving people clear details about how their data will be used and their rights

Answer

Getting consent only from the person in charge of the data

Answer

Assuming people agree if they don't say no

Answer

Using consent forms that are hard to understand

Question 2

What is a potential societal concern associated with using biased machine learning systems in healthcare?

Accepted Answer

Unfair treatment

Answer

Increased access to healthcare

Answer

Improved health outcomes

Answer

Reduced healthcare costs

Question 3

Which of these is NOT a common method for addressing bias in machine learning algorithms?

Accepted Answer

Subjective human judgment

Answer

Data preprocessing

Answer

Post-processing

Answer

Algorithmic fairness constraints

Question 4

Which ethical principle in machine learning emphasizes the importance of avoiding harm?

Accepted Answer

Non-maleficence

Answer

Maximize benefit

Answer

Respect autonomy

Answer

Promote justice

Question 5

Which type of bias occurs when a machine learning algorithm is influenced by prior beliefs or expectations?

Accepted Answer

Confirmation bias

Answer

Selection bias

Answer

Availability bias

Answer

Sampling bias

Question 6

What is the primary function of the Equal Employment Opportunity Commission (EEOC) in relation to machine learning?

Accepted Answer

To investigate and enforce laws prohibiting discrimination in employment practices, including the use of machine learning.

Answer

To regulate the specific use of machine learning algorithms in hiring and other employment decisions.

Answer

To provide guidance and resources on how to use machine learning in a fair and non-discriminatory manner.

Answer

To train employers and organizations on how to avoid bias and discrimination in their machine learning practices.

Question 7

What is the primary purpose of an algorithmic audit in ensuring fairness in machine learning?

Accepted Answer

To identify and address potential biases and discriminatory practices within machine learning algorithms.

Answer

To ensure that machine learning algorithms are fully compliant with all applicable laws and regulations.

Answer

To provide comprehensive documentation for the development and deployment of machine learning algorithms.

Answer

To train machine learning engineers on ethical and fair practices in developing and deploying algorithms.

Question 8

What is the concept of 'proxy bias' in machine learning?

Accepted Answer

When a machine learning algorithm uses a proxy variable to stand in for a protected characteristic, such as race or gender.

Answer

When a machine learning algorithm is biased against a particular group of people even though it does not have direct information about their protected characteristics.

Answer

When a machine learning algorithm is used to make decisions that have a negative impact on a particular group of people.

Answer

When a machine learning algorithm is trained on a biased dataset.

Question 9

Which of the following is not a type of bias that can impact machine learning algorithms?

Accepted Answer

Confirmation bias

Answer

Selection bias

Answer

Statistical bias

Answer

Sampling bias

Question 10

In the context of algorithmic fairness, which group is legally shielded from biased treatment?

Accepted Answer

Individuals belonging to a protected class

Answer

Individuals disproportionately misclassified by algorithms

Answer

Individuals who benefit the most from machine learning applications

Answer

Individuals who are underrepresented in machine learning datasets

Question 11

Which ethical principle emphasizes the importance of creating fair and unbiased machine learning algorithms?

Accepted Answer

Justice

Answer

Autonomy

Answer

Non-maleficence

Answer

Beneficence

Question 12

What is the term used to describe an algorithm that performs well on the training data but poorly on unseen data?

Accepted Answer

Overfitting

Answer

Underfitting

Answer

Discrimination

Answer

Fairness

Question 13

Which ethical principle emphasizes the importance of fairness and justice in machine learning systems?

Accepted Answer

Justice

Answer

Beneficence

Answer

Non-maleficence

Answer

Autonomy

Question 14

Which of the following is a common source of bias in machine learning algorithms?

Accepted Answer

Training data that is not representative of the intended population

Answer

The complexity of the mathematical models used

Answer

The lack of human supervision during model development

Question 15

What is the primary goal of fairness in machine learning?

Accepted Answer

To ensure that algorithms treat all individuals equitably and without discrimination

Answer

To minimize the computational cost of training models

Answer

To enhance the precision of predictions

Question 16

What is the key distinction between fairness and equality in machine learning?

Accepted Answer

Fairness aims to handle individuals justly, while equality aims for similar outcomes

Answer

Fairness is a subjective concept, whereas equality is objective

Answer

Fairness is more significant than equality in machine learning

Question 17

How does transparency contribute to promoting fairness in machine learning?

Accepted Answer

It enables stakeholders to comprehend how algorithms make decisions

Answer

It mitigates the risk of bias in algorithms

Answer

It improves prediction accuracy

Question 18

What is the ethical challenge faced by machine learning practitioners in addressing fairness and discrimination?

Accepted Answer

Balancing the necessity for precise predictions with the imperative to avert unjust outcomes

Answer

Determining suitable fairness metrics

Answer

Deciding whether to utilize machine learning algorithms entirely

Question 19

Why is public understanding and education crucial for promoting fairness in machine learning?

Accepted Answer

It empowers individuals to comprehend and advocate for their rights

Answer

It decreases the demand for governmental regulation

Answer

It expedites the development of innovative machine learning algorithms

Question 20

What is the duty of machine learning practitioners in upholding fairness and preventing discrimination?

Accepted Answer

To design, implement, and evaluate algorithms responsibly and ethically

Answer

To educate the public about the risks of machine learning

Answer

To lobby for legal regulation of machine learning

Question 21

How do human-in-the-loop approaches enhance fairness in machine learning?

Accepted Answer

Incorporate human judgment and feedback into model development and deployment

Answer

Reduce errors in training data

Answer

Ensure compliance with laws and ethics

Question 22

Which of the following is NOT a common type of bias in machine learning algorithms?

Accepted Answer

Randomness

Answer

Algorithmic bias

Answer

Label bias

Answer

Sampling bias

Question 23

What is the primary purpose of cross-validation in machine learning?

Accepted Answer

To evaluate the predictive performance of a model on unseen data.

Answer

To tune the hyperparameters of a model.

Answer

To clean and preprocess the training data.

Question 24

Which of the following is a metric commonly used to assess the fairness of a machine learning model?

Accepted Answer

Disparate impact

Answer

Recall

Answer

F1-Score

Answer

Precision

Question 25

Which of the following is a potential negative consequence of algorithmic bias?

Accepted Answer

Unfair or discriminatory decision-making

Answer

Reduced model accuracy

Answer

Increased data leakage

Answer

Higher computational cost

Question 26

Why is interpretability important when considering the fairness of a machine learning model?

Accepted Answer

It enables identification and understanding of potential biases influencing model predictions.

Answer

It simplifies model implementation and reduces maintenance overhead.

Answer

It improves the model's accuracy and reduces computational cost.

Question 27

Which of the following laws aims to address the ethical use of artificial intelligence?

Accepted Answer

Fairness in Artificial Intelligence Act (FAIA)

Answer

Health Insurance Portability and Accountability Act (HIPAA)

Answer

Americans with Disabilities Act (ADA)

Answer

Gramm-Leach-Bliley Act (GLBA)

Question 28

What is the key distinction between overt and covert discrimination?

Accepted Answer

Overt discrimination is intentional, while covert discrimination is unintentional.

Answer

Overt discrimination targets individuals, while covert discrimination affects groups.

Answer

Overt discrimination is illegal, while covert discrimination is legal.

Question 29

Why is it crucial to consider fairness and discrimination in the development and deployment of machine learning models?

Accepted Answer

To ensure the ethical and responsible use of machine learning technology.

Answer

To guarantee highly accurate and efficient models.

Answer

To minimize the development costs associated with machine learning.

Question 30

How can machine learning contribute to the promotion of fairness and the reduction of discrimination?

Accepted Answer

By identifying and eliminating biases from decision-making processes

Answer

By automating decision-making tasks

Answer

By enhancing the speed and efficiency of decision-making processes

Question 31

What is a key challenge associated with using machine learning to combat discrimination?

Accepted Answer

The difficulty in detecting and quantifying discrimination

Answer

The high cost of implementing machine learning solutions

Answer

The shortage of qualified machine learning professionals

Question 32

What is the concept of "protected characteristics" in the context of fair machine learning?

Accepted Answer

Attributes such as race, gender, and religion that are legally protected from discrimination

Answer

Attributes that are easily measurable

Answer

Attributes that are directly relevant to the decision-making task

Question 33

Which of the following exemplifies an unfair decision-making process?

Accepted Answer

A hiring algorithm that favors male applicants over female applicants

Answer

A loan approval algorithm that considers credit score

Answer

A medical diagnosis algorithm that utilizes patient health data

Question 34

What is the significance of transparency in the context of fair machine learning?

Accepted Answer

It enables stakeholders to comprehend decision-making processes and identify potential biases

Answer

It reduces the likelihood of discrimination by making algorithms more predictable

Answer

It enhances the accuracy of machine learning models by providing additional information

Question 35

Which of the following represents a best practice for developing fair machine learning models?

Accepted Answer

Involving stakeholders from diverse backgrounds in the design and evaluation stages

Answer

Utilizing the largest possible training dataset

Answer

Selecting the most complex machine learning algorithm

Question 36

What is the primary objective of fair machine learning?

Accepted Answer

Developing machine learning systems that make equitable and non-discriminatory decisions

Answer

Eliminating bias from machine learning algorithms entirely

Answer

Maximizing the accuracy of machine learning models

Question 37

Which of the following is NOT a potential bias in machine learning algorithms?

Accepted Answer

Non-determinism

Answer

Overfitting

Answer

Confirmation bias

Answer

Sampling bias

Question 38

What is the primary goal of fairness in machine learning?

Accepted Answer

To ensure that algorithms make decisions that are fair and unbiased.

Answer

To improve the accuracy of algorithms.

Answer

To reduce the computational complexity of algorithms.

Question 39

What is the ethical principle that requires algorithms to be accountable for their decisions?

Accepted Answer

Transparency

Answer

Beneficence

Answer

Autonomy

Answer

Justice

Question 40

Which federal law in the United States prohibits discrimination in employment, housing, and public accommodations?

Accepted Answer

Civil Rights Act of 1964

Answer

Equal Employment Opportunity Act

Answer

Fair Housing Act

Answer

Americans with Disabilities Act

Question 41

What is the term for the belief that one's own group or culture is superior to others?

Accepted Answer

Ethnocentrism

Answer

Xenophobia

Answer

Bigotry

Answer

Racism

Question 42

How can machine learning algorithms contribute to the perpetuation of discrimination?

Accepted Answer

By amplifying existing biases in the data used to train them.

Answer

Algorithms can only discriminate if they are explicitly programmed to do so.

Answer

Algorithms can create entirely new forms of discrimination.

Question 43

What is the main purpose of the Equal Employment Opportunity Commission (EEOC)?

Accepted Answer

To enforce federal laws prohibiting employment discrimination.

Answer

To provide legal advice to employers.

Answer

To investigate complaints of discrimination.

Question 44

What is the primary purpose of using cross-validation in machine learning?

Accepted Answer

To evaluate the performance of a model on unseen data and mitigate overfitting.

Answer

To train a model with different hyperparameters.

Answer

To improve the accuracy of a model without overfitting.

Question 45

In the context of machine learning, what is the key distinction between fairness and discrimination?

Accepted Answer

Fairness pertains to the absence of bias, while discrimination refers to the unfair or prejudicial treatment of individuals based on protected attributes.

Answer

Fairness is related to the interpretability of a model, while discrimination is related to the accuracy of a model.

Question 46

Which of the following is NOT a protected characteristic under the United States Fair Housing Act?

Accepted Answer

Gender identity

Answer

National origin

Answer

Race

Answer

Religion

Question 47

What is the primary responsibility of the Equal Employment Opportunity Commission (EEOC)?

Accepted Answer

Enforcing federal laws that prohibit discrimination in employment.

Answer

Providing legal representation to victims of discrimination.

Answer

Investigating allegations of workplace harassment.

Question 48

Which legal defense can be invoked to justify a practice that has a disparate impact on a protected group?

Accepted Answer

Business necessity

Answer

Statistical anomaly

Answer

Affirmative action

Answer

Good faith

Question 49

Which of the following is NOT a potential remedy for discrimination?

Accepted Answer

Criminal prosecution

Answer

Back pay

Answer

Injunctions

Answer

Reinstatement

Question 50

Why are ethical considerations crucial in the development and implementation of machine learning algorithms?

Accepted Answer

To ensure algorithms are fair, unbiased, respectful of individual rights and freedoms, and aligned with societal values.

Answer

Ethical considerations are only relevant for large-scale machine learning projects.