Artificial Intelligence Fundamentals GOAA Practice Questions (Page 9)

Questions 41–45

1. 41foundation · easy

   What role do activation functions play in a neural network?

   • AThey introduce non-linearity, allowing the network to learn complex patterns.
   • BThey reduce the number of layers in the network.
   • CThey initialize the weights of the network.
   • DThey store the training data for later retrieval.
   Reveal Solution Discussion Select an answer first
2. 42application · medium

   A company uses an AI chatbot for customer support. The chatbot sometimes provides confident but incorrect answers to user questions. What is the primary limitation of AI systems that this scenario illustrates?

   • ALack of true understanding and reasoning
   • BInsufficient training data
   • COverfitting to the training data
   • DLack of computational power
   Reveal Solution Discussion Select an answer first
4. 43foundation · easy

   In supervised learning, what does the model learn from?

   • AData that has both input features and corresponding correct output labels
   • BData with no labels, finding hidden structure on its own
   • COnly the input data, without any feedback
   • DUnlabeled data and a reward signal
   Reveal Solution Discussion Select an answer first
5. 44application · medium

   A security operations center (SOC) wants to deploy an AI system that flags phishing emails. The team has a large dataset of emails labeled 'phishing' or 'benign'. They want the model to learn patterns from these labels and then generalize to new emails. Which machine learning approach best fits this requirement?

   • AReinforcement learning with a reward function
   • BSupervised learning with a classification algorithm
   • CUnsupervised learning with clustering
   • DDeep learning with a generative adversarial network
   Reveal Solution Discussion Select an answer first
6. 45application · easy

   A retail company wants to segment its customers into groups based on purchasing behavior to tailor marketing campaigns. They have no pre-existing labels for customer segments. Which AI approach should they use?

   • ADeep learning with a convolutional neural network
   • BSupervised learning with classification
   • CUnsupervised learning with clustering
   • DReinforcement learning with a reward function
   Reveal Solution Discussion Select an answer first