Define Machine Learning Concepts for Data Science Workloads SNOWPRO-ADVANCED-DATA-SCIENTIST Practice Questions (Page 2)

Questions 6–10

1. 6expert · hard

   A model is underfitting. The data scientist has already increased model complexity and added features, but performance is still poor. What is the most likely remaining cause?

   • AThe model is overfitting despite low training error.
   • BThe learning rate is too high.
   • CThe validation set is too small.
   • DThe data may contain noise or irrelevant features that confuse the model.
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   Which strategy is commonly used to mitigate overfitting in a machine learning model?

   • AReducing the number of features without validation
   • BRemoving regularization
   • CUsing more training data
   • DIncreasing model complexity
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A data scientist is training a neural network. The training loss decreases steadily, but the validation loss starts to increase after 20 epochs. What is the best course of action?

   • AIncrease the learning rate to escape the local minimum.
   • BContinue training for more epochs to see if validation loss decreases later.
   • CRemove the validation set and train on all data.
   • DStop training early and use the model from epoch 20.
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A team is building a model to predict rare events. The dataset is highly imbalanced (1% positive class). They are using accuracy as the evaluation metric and the model achieves 99% accuracy. Which of the following is the most important issue?

   • AThe model is overfitting because accuracy is too high.
   • BThe model is underfitting because it cannot predict the rare class.
   • CAccuracy is misleading because a model that always predicts the majority class would achieve 99% accuracy.
   • DThe validation set is too small.
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   A data scientist is building a model to predict whether a customer will churn, using historical data where each customer is labeled as 'churned' or 'not churned'. Which type of machine learning is this?

   • AReinforcement learning
   • BSupervised learning
   • CUnsupervised learning
   • DSemi-supervised learning
   Reveal Solution Discussion Select an answer first