Is it possible to completely eliminate all errors in a machine learning model?

Group of answer choices

No$,$ unavoidable error $($irreducible error$)$ exists in all machine learning problems.

No$,$ by increasing the model's complexity, all errors can be reduced to negligible levels.

Yes, with careful feature engineering and hyperparameter tuning, all errors can be eliminated.

Yes, given enough data, the model can learn to handle any complexity in the dataset.