Multiple Choice Questions on Fairness in AI Algorithms for AP CSP

📚 Quick Study Guide

🔍 Fairness in AI: Addresses biases in algorithms, ensuring equitable outcomes across different groups. ⚖️ Bias Detection: Identifying and mitigating biases in training data and algorithm design. 🤖 Types of Bias: Includes historical bias (reflecting past discrimination), representation bias (skewed data samples), and measurement bias (inaccurate data collection). 📈 Evaluation Metrics: Assessing fairness using metrics like disparate impact (equal outcomes across groups) and equal opportunity (equal chances of positive outcomes). 🛡️ Mitigation Techniques: Methods to reduce bias, such as re-weighting training data, adversarial debiasing, and fairness-aware algorithms. 🧑‍🤝‍🧑 Ethical Considerations: Emphasizing transparency, accountability, and the potential social impact of AI systems. 📝 Legal Frameworks: Understanding relevant regulations and guidelines related to AI fairness and non-discrimination.

🧪 Practice Quiz

1. Which of the following best describes fairness in AI algorithms?
   1. A) Achieving 100% accuracy for all users.
   2. B) Ensuring equitable outcomes across different demographic groups.
   3. C) Minimizing computational complexity.
   4. D) Maximizing profit for the developing company.
2. What is historical bias in AI?
   1. A) Bias introduced during the coding phase of AI development.
   2. B) Bias reflecting past societal discrimination present in training data.
   3. C) Bias caused by using outdated hardware to run AI models.
   4. D) Bias resulting from algorithms being trained for too long.
3. Which of the following is an example of representation bias?
   1. A) An AI model that always gives the same answer.
   2. B) A dataset where one demographic group is significantly underrepresented.
   3. C) An algorithm designed to favor a specific outcome.
   4. D) A situation where the algorithm is too complex to understand.
4. What does 'disparate impact' refer to in the context of AI fairness?
   1. A) The degree to which an algorithm is difficult to understand.
   2. B) Unequal outcomes produced by an algorithm across different groups.
   3. C) The amount of computational resources required to run an algorithm.
   4. D) The speed at which an algorithm can process data.
5. Which technique involves adjusting the importance of different data points to reduce bias?
   1. A) Data normalization.
   2. B) Re-weighting training data.
   3. C) Feature selection.
   4. D) Model compression.
6. Why is transparency important in AI algorithms?
   1. A) To ensure the algorithms run faster.
   2. B) To hide the complexities of the algorithm from users.
   3. C) To understand how decisions are made and identify potential biases.
   4. D) To reduce the cost of developing the algorithm.
7. What is a key ethical consideration when deploying AI systems?
   1. A) Maximizing computational efficiency.
   2. B) Ignoring potential biases to achieve faster results.
   3. C) Evaluating the potential social impact and ensuring accountability.
   4. D) Keeping the algorithm a secret to maintain a competitive edge.