Quick Check 11.7 – Data 100, Summer 2020
Sign in to Google to save your progress. Learn more
Is the optimal value of theta when using squared loss always greater than the optimal value of theta when using absolute loss? *
Consider the toy dataset [1, 3, 5, 8, 13]. When using the constant model and squared loss, the optimal theta is 6, and the MSE is 17.6. When using the constant model and absolute loss, the optimal theta is 5, and the MAE is 17. Does this mean that thetahat = 5 is more optimal than thetahat = 6? *
Which loss function is more sensitive to outliers? *
Submit
Clear form
This form was created inside of UC Berkeley. Report Abuse
Google Forms