MLA-C01 Question 174
Select 3You are a Machine Learning Engineer developing a model on Amazon SageMaker to predict customer churn. Despite having a high training accuracy, your model exhibits poor performance on the validation dataset. Which of the following steps can you take to improve the model's performance?
- A
Increase the size of the training dataset by collecting more labeled data.
- B
Reduce the model complexity by decreasing the number of layers in the neural network.
- C
Enable automatic model tuning (hyperparameter optimization) in SageMaker.
- D
Add more features without checking their relevance to the target variable.
- E
Apply regularization techniques such as L1 or L2 regularization to the model.
Show answer and explanation
Correct answers: A, C, E
Explanation
To improve model performance on the validation dataset, focus on addressing overfitting and optimizing the model's hyperparameters. Increasing the training dataset size, using automatic model tuning, and applying regularization are effective strategies. Reducing model complexity may lead to underfitting, and adding irrelevant features can introduce noise, both of which are counterproductive.
- A. Correct.
Increasing the size of the training dataset can help the model generalize better, reducing overfitting and improving validation performance.
- B. Incorrect.
Reducing the model complexity can lead to underfitting, which may not address the issue of poor performance on the validation dataset.
- C. Correct.
Using SageMaker's automatic model tuning can help optimize hyperparameters, such as learning rate and batch size, to improve the model's performance on validation data.
- D. Incorrect.
Adding more features without evaluating their relevance can introduce noise and may negatively impact the model's performance.
- E. Correct.
Applying regularization techniques like L1 or L2 helps reduce overfitting by penalizing large weights in the model, which can improve generalization.