Machine Learning and Neural Networks AIE Practice Questions (Page 6)

Questions 26–30

1. 26foundation · easy

   Which statement accurately describes a deep neural network?

   • AIt has only one hidden layer but many neurons in that layer.
   • BIt requires a GPU to be trained.
   • CIt has multiple hidden layers between the input and output layers.
   • DIt uses only linear activation functions.
   Reveal Solution Discussion Select an answer first
2. 27application · easy

   A team is building a model to recognize handwritten digits. They start with a simple neural network with one hidden layer, but the accuracy is insufficient. They decide to add more hidden layers to capture complex patterns. What is this approach called?

   • ADeep learning, because the network has multiple hidden layers
   • BDimensionality reduction, because the network reduces the input size
   • CUnsupervised learning, because the network finds patterns without labels
   • DReinforcement learning, because the network learns from rewards
   Reveal Solution Discussion Select an answer first
4. 28application · medium

   A game development studio is training an AI agent to play a new board game. The agent makes moves, receives a reward when it wins and a penalty when it loses, and gradually improves its strategy. Which components are essential for this reinforcement learning setup?

   • AA clustering algorithm, a distance metric, and a number of clusters
   • BA fixed set of rules, a decision tree, and a pruning strategy
   • CAn agent, an environment, a reward signal, and a policy that maps states to actions
   • DA labeled dataset, a loss function, and backpropagation
   Reveal Solution Discussion Select an answer first
5. 29foundation · easy

   During backpropagation, what information is propagated backward through the network?

   • AThe activation values of each neuron to be recalculated.
   • BThe final predictions to be compared with the targets.
   • CThe gradient of the loss with respect to each weight.
   • DThe original input data to be reprocessed.
   Reveal Solution Discussion Select an answer first
6. 30foundation · easy

   In reinforcement learning, what is the role of the policy?

   • AIt defines the set of all possible states in the environment.
   • BIt provides the reward signal after each action.
   • CIt defines the strategy that the agent uses to decide which action to take in a given state.
   • DIt stores all past experiences for the agent to replay.
   Reveal Solution Discussion Select an answer first