A machine learning (ML) specialist is building a credit score model for a financial institution. The ML specialist has collected data for the previous 3 years of transactions and third-party…

Question

Options

• AIncrease the number of passes on the existing training data. Perform more hyperparameter tuning.
• BIncrease the amount of regularization. Use fewer feature combinations.
• CAdd new domain-specific features. Use more complex models.
• DUse fewer feature combinations. Decrease the number of numeric attribute bins.
• EDecrease the amount of training data examples. Reduce the number of passes on the existing

• A
  71% (44)
• B
  16% (10)
• D
  3% (2)
• E
  10% (6)

Explanation

Increasing the number of passes (or epochs) on the training data allows the model to learn more from the data, especially if the initial training was underfitting. Hyperparameter tuning can help find the optimal settings (such as learning rate, batch size, or regularization strength) that can improve model accuracy. Adding new, domain-specific features (i.e., features that capture important domain knowledge) can greatly improve the model's ability to generalize and capture patterns in the data. Using more complex models can also increase the model's ability to fit to the training data, which might help if the initial model was too simple (underfitting). Increasing regularization and using fewer feature combinations may help in reducing overfitting but won't address underfitting, which seems to be the issue here. Using fewer feature combinations and decreasing numeric bins would simplify the model, but this could make the problem worse if underfitting is already happening. Decreasing training data and reducing passes would likely worsen the performance by limiting the model's exposure to the data and reducing the amount of learning.

No community discussion yet for this question.