L1 regularization, which adds a penalty term to the loss function based on the absolute values of the model parameters.

L2 regularization, which adds a penalty term to the loss function based on the squared values of the model parameters.

Both A and B, as Spark MLlib supports both L1 and L2 regularization techniques for linear regression models.

None of the above, as regularization techniques are not applicable to linear regression models in Spark MLlib.