NCA-GENL Question 32
Select 3You are training a machine learning model to classify images of fruits. During the development process, you realize that the model's performance on the validation set is significantly worse than on the training set. Which of the following steps could help improve the model's validation performance?
- A
Increase the size of the training dataset by augmenting the images.
- B
Use cross-validation to better evaluate model performance.
- C
Reduce the complexity of the model to prevent overfitting.
- D
Train the model for more epochs to improve training accuracy.
- E
Add more features to the dataset to improve model learning.
Show answer and explanation
Correct answers: A, B, C
Explanation
To improve validation performance when the model is overfitting, strategies like increasing the training dataset size, using cross-validation for better evaluation, and reducing model complexity are effective. These approaches address overfitting by enhancing generalization or providing a more accurate assessment of the model's performance on unseen data.
- A. Correct.
Increasing the size of the training dataset, especially through data augmentation, helps the model generalize better and reduces overfitting, which can improve validation performance.
- B. Correct.
Using cross-validation provides a more robust estimate of model performance on unseen data, which can help identify and address overfitting or data leakage issues.
- C. Correct.
Reducing model complexity (e.g., fewer layers or parameters) mitigates overfitting by restricting the model's capacity to memorize training data, thereby improving validation performance.
- D. Incorrect.
Training the model for more epochs without addressing overfitting will likely lead to further overfitting, worsening validation performance.
- E. Incorrect.
Adding more features without ensuring their relevance or quality can lead to feature redundancy or noise, which might degrade model performance.