Quick Check 14.7 – Data 100, Summer 2020

JavaScript isn't enabled in your browser, so this file can't be opened. Enable and reload.

In Lecture 12, we established that the RMSE of a model cannot go up by adding more features. Why do the results studied in this lecture – where the model with the most features had the highest error – not contradict this result? *

Answer using a sentence. A correct answer will appear after you submit.

Why do we want to avoid overfitting? *

To reduce the number of model parameters so that the amount of processing power required to make predictions is lower

So that our model is well-suited for making good predictions on unseen data, and hasn't just "learned the noise"

To ensure that our design matrix has full column rank

To ensure that we are still able to add features to our design matrix

Submit

Clear form

Never submit passwords through Google Forms.

This form was created inside of UC Berkeley.

Does this form look suspicious? Report

Forms