Which of the following is true about model selection algorithms?

Question $8$ options:

Best subset regression considers all possible combinations of variables. Therefore it guarantees the best performing combination of variables.

Forward selection and backward selection algorithms will choose the same variables as long as you choose the same alpha.

Forward selection and Stepwise selection will return same in multiple regression analysis.

Backward selection algorithm ends up coming up with the intercept alone in the very last step.