Question 3What is overfitting in machine learning?a) Overfitting occurs when a model has high complexity and captures both information and noise in the training data.b) Overfitting occurs when a model has poor performance on the training data.c) Overfitting is indicated when a model has good performance on the training dataset but relatively poor performance on the testing dataset.d) Overfitting occurs when a model has good performance on the test data.Answer choicesSelect only one optionREVISITa onlya & da & ca, c & d

Question 7What method can you use to minimize overfitting of a machine learning model?1 pointIncrease the variance of your training data.Tune the hyperparameters of your model using cross-validation.Choose the hyperparameters that maximize goodness of fit on your training data.Decrease the variance of your test data.

Question 18Which statement is true about overfitting?1 pointThe model is too flexible and fits the noise rather than the function.If the model is noisy, you need a low-order polynomial so you don’t overfit the data.The higher the order of the polynomial, the less overfitting occurs.If a model is overfit with the training data it will also overfit the testing data.

Which of the following is NOT a typical method to improve an overfitting machine learning model?Add more dataSelect more featuresSelect a simpler algorithmImprove feature engineering

Overfitting occurs when a model:Question 7AnswerA.Has high bias and low varianceB.Has low bias and high varianceC. Performs well on unseen dataD. Fails to capture the underlying patterns in the data

6. Which of the following can we use to solve the problem of “overfitting” in a neural network?我們可以使用以下哪項解決神經網路中的「過度擬合」問題？Regularization 正則項／懲罰項Activation function 激勵函數Epoch 訓練次數All of the above options 以上選項皆可