1. A stakeholder asks how the AI model will adapt to changes in the business environment. Which response best addresses their concern?
2. When working with an imbalanced dataset, which resampling technique involves creating synthetic samples of the minority class?
3. How does increasing the size of a dataset typically affect a machine learning model?
4. Which architecture is typically used to deploy machine learning models as a service?
5. What is the main ethical issue with using historical data that reflects past discriminatory practices in training an ML model?
6. Which evaluation metric would you choose to minimize if your model's errors should not be overly penalized by large deviations?
7. In a production environment, what is the purpose of implementing canary deployment for machine learning models?
8. A stakeholder asks how AI models are evaluated for success. Which metric is most important to emphasize?
9. A marketing team is using AI to personalize email campaigns. What challenge might arise, and how can it be mitigated?
10. Which optimization algorithm is known for combining momentum and adaptive learning rates to achieve better performance in deep learning?
11. What does the Area Under the Receiver Operating Characteristic Curve (AUC-ROC) measure in model evaluation?
12. In the context of maintaining a machine learning model in production, what is the primary purpose of model retraining?
13. Which of the following is a benefit of using a model monitoring system in production?
14. Which technique can improve the reliability of model performance estimates when working with small datasets?
15. Which practice is recommended for mitigating the risk of model inversion attacks?
16. A stakeholder asks how the AI model can handle unexpected inputs or scenarios not covered in the training data. Which response best explains the situation?
17. Why is it important to apply the same transformation consistently to both training and test datasets?
18. In the context of reinforcement learning, what does the term "reward" refer to?
19. What is the most effective way to address ethical concerns related to fairness in AI models used for loan approvals?
20. Which challenge is most commonly associated with deploying machine learning models in real-time systems?
21. How can companies ensure that their AI models do not unintentionally perpetuate discrimination in hiring practices?
22. Which of the following is a key ethical concern when developing AI models for predictive policing?
23. Which of the following is a key characteristic of a Convolutional Neural Network (CNN) that makes it effective for image classification tasks?
24. Which of the following techniques is most likely to reduce the training time of a deep neural network?
25. A government agency needs to identify fraudulent transactions in a large dataset. What is the best strategy to balance the detection of fraud with minimizing false positives?
26. How does target encoding differ from one-hot encoding in handling categorical data?
27. When operationalizing a machine learning model, what is the significance of monitoring model latency in production?
28. Which ethical concern is associated with AI in surveillance systems?
29. In video data processing, what is a common method to reduce the amount of data without losing critical information?
30. Which of the following methods is effective for protecting a machine learning pipeline from insider threats?
31. What is the primary function of a neural network in artificial intelligence?
32. Which of the following is an effective approach to reduce the computational cost of hyperparameter tuning for deep learning models?
33. An AI model is used to forecast demand for energy in a smart grid. What is a key technical challenge, and how can it be addressed?
34. Which data format is most suitable for storing high-dimensional text data?
35. Which cross-validation technique is often used to ensure that each data point is used for validation exactly once?
36. What is the primary reason for implementing explainability in a machine learning model deployed in a high-stakes decision-making environment?
37. What is the primary ethical concern of using AI for facial recognition in public surveillance?
38. What is the most significant risk of training a model on a small, low-quality dataset?
39. A large-scale commercial enterprise aims to use AI for personalized marketing. How can the company ensure that its AI models respect customer autonomy while delivering personalized experiences?
40. Which of the following is a challenge associated with using deep learning models in AI?
41. Why is it important to consider the context of the application when choosing an evaluation metric for a machine learning model?
42. Which strategy is most effective in ensuring that an AI system aligns with ethical standards during its development?
43. In the context of neural networks, what is backpropagation used for?
44. What role does feature engineering play in ensuring the ethical use of AI in decision-making processes?
45. In feature engineering, what is the primary purpose of using polynomial features?
46. Which of the following is a common strategy to detect and mitigate bias in machine learning models?
47. How can organizations ensure that their AI systems do not exacerbate existing inequalities in society?
48. What is the role of a data pipeline in the deployment of machine learning models?
49. What is the primary goal of using cross-validation in model training?
50. In continuous integration/continuous deployment (CI/CD) pipelines for machine learning, what is the role of automated testing?
51. In reinforcement learning, what is the “exploration-exploitation trade-off”?
52. In the context of transfer learning, which of the following scenarios would most likely lead to negative transfer?
53. In a financial institution, which business case would most likely support the adoption of speech recognition technology?
54. Which factor is most likely to negatively impact model generalization?
55. Which of the following is a benefit of using the Rectified Linear Unit (ReLU) activation function in deep neural networks?
56. In video data processing, what is the purpose of extracting key frames?
57. How can businesses best manage the risk of deploying machine learning models that make consequential decisions for individuals?
58. Which of the following is a critical consideration for ensuring transparency in a machine learning model used in a legal context?
59. Why is it important to consider the size of a dataset when selecting a machine learning algorithm?
60. In which scenario is reinforcement learning most applicable?
61. What is the primary purpose of using a security information and event management (SIEM) system in a machine learning pipeline?
62. What is a common business risk associated with improper feature selection in a machine learning model?
63. What is overfitting in the context of machine learning models?
64. Which of the following challenges is typically associated with the deployment of deep learning models in production?
65. Which machine learning deployment pattern involves serving multiple models that each handle specific subsets of data?
66. Which concept refers to the ability to deploy machine learning models in different environments without altering the code?
67. A company uses AI to monitor and predict machine failures in its manufacturing process. What would be a critical step to ensure the reliability of these predictions?
68. Which technique is used to ensure that a machine learning model deployed in production can handle a sudden increase in traffic securely?
69. During a presentation, a stakeholder expresses concern about the AI model making decisions without human oversight. How would you best address this concern?
70. What is the primary advantage of using min-max normalization in neural networks?
71. What does a large gap between training and validation accuracy typically suggest about a model?
72. Why is it crucial to apply transformations like standardization or normalization in machine learning workflows?
73. Which of the following is a common challenge when applying transfer learning to a new domain?
74. Which of the following business cases would justify the implementation of speech recognition in a call center?
75. What is a potential drawback of using min-max normalization on numerical data?
76. Which technique is used to ensure that a deployed machine learning model generalizes well to new, unseen data?
77. Why is transparency in AI decision-making processes crucial for maintaining public trust?
78. Why is it critical to keep the test set separate and untouched during the model training process?
79. What is the primary purpose of using a validation dataset in machine learning?
80. Which of the following is the most significant ethical concern when deploying an AI model in a healthcare setting?