The correct answer is: (iii) Models which overfit are more likely to have high variance.
Bias and variance are two key concepts in machine learning. Bias refers to the error made by a model due to its systematic underestimation or overestimation of the true value. Variance refers to the error made by a model due to its sensitivity to small changes in the training data.
Overfitting occurs when a model learns the training data too well and does not generalize well to new data. This can happen when a model is too complex or when the training data is too small. Overfitting can be avoided by using regularization techniques, such as L2 regularization or dropout.
Models which overfit are more likely to have high variance because they are too sensitive to small changes in the training data. This means that they will perform well on the training data but poorly on new data.
Here is a more detailed explanation of each option:
(i) Models which overfit are more likely to have high bias. This is not true. Overfitting is caused by a model learning the training data too well, which leads to high variance. High bias is caused by a model not learning the training data well enough, which leads to low variance.
(ii) Models which overfit are more likely to have low bias. This is also not true. Overfitting is caused by a model learning the training data too well, which leads to high variance. Low bias is caused by a model learning the training data well enough, which leads to low variance.
(iii) Models which overfit are more likely to have high variance. This is true. Overfitting is caused by a model learning the training data too well, which leads to high variance.
(iv) Models which overfit are more likely to have low variance. This is not true. Overfitting is caused by a model learning the training data too well, which leads to high variance. Low variance is caused by a model learning the training data well enough, which leads to low variance.