Which of the following are true about overfitting? regularizing helps reduce overfitting Having a low training error and high testing error is a good indicator of overfitting Having a high training and testing error is a good indicator of overfitting Having too simple of a model causes overfitting Overfitting is due to fitting to the noise of the data. Overfitting can arise due to having too many degrees of freedom Polynomial regression using ordinary least squares without regularization requires which of the following hyperparamters to be set? lambda (regularization term) weights alpha (learning rate) degree Which Linear Regression algorithm would we want to use if we had a training dataset with millions of data samples? None of the above Ordianry Least Sqaures Least Mean Squares The following polynomial has how many degrees? 22+2x+3x3+9x4