AWS Certified Machine Learning Specialty · Free Practice Question Medium
Question 21
After training a SageMaker Linear Learner model over a training dataset, the data science team observed that it achieved high accuracy on the training data, but had low accuracy on the test data.
As an AWS ML specialist, which of the following techniques are likely to help resolve this problem? (Select three)
-
A
Use more training data
-
B
Add regularization to the model
-
C
Remove regularization from the model
-
D
Use less training data
-
E
Use more features in the model
-
F
Use less features in the model
Reveal correct answers
Correct answers: A, B, F
Explanation
Correct options:
Use more training data
Add regularization to the model
Use less features in the model
Your model is underfitting the training data when the model performs poorly on the training data. This is because the model is unable to capture the relationship between the input examples (often called X) and the target values (often called Y). Your model is overfitting your training data when you see that the model performs well on the training data but does not perform well on the evaluation data. This is because the model is memorizing the data it has seen and is unable to generalize to unseen examples.
As the model has high accuracy on the training data but low accuracy on the test data, it suggests that the model is overfitting. When a model is overfitting then adding more training data, adding regularization or using less features can help in addressing the underlying problem.
Incorrect options:
Use more features in the model
Remove regularization from the model
Use less training data
These three options contradict the explanation provided above, so these options are incorrect.
References:
https://docs.aws.amazon.com/machine-learning/latest/dg/model-fit-underfitting-vs-overfitting.html
https://zhu45.org/posts/2017/Jul/21/andrew-ngs-ml-week-06-11/#overfitting-vs-underfitting
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
