Explanation:
L1 regularization is preferred when you need to emphasize more influential features by shrinking less important ones to zero. L2 regularization is more suitable when all features have a similar impact on the output. For more details, refer to: Google's Machine Learning Crash Course on Regularization.
Ultimate access to all questions.
No comments yet.
When training a neural network to predict housing prices using a dataset that includes latitude and longitude, what is the most effective method to account for the influence of location on price?
A
Directly inputting latitude and longitude as vectors into the neural network
B
Generating a numeric column by crossing latitude and longitude features
C
Crossing latitude and longitude features, bucketizing at the minute level, and applying L1 regularization during optimization
D
Crossing latitude and longitude features, bucketizing at the minute level, and applying L2 regularization during optimization