[samihaq]

Hi,
I have a very basic question.
Since we have h(x) = b as hypothesis set then as it is just a constant so its value does not change with different value of 'x' but can change for different h's, i am confused that what would be the learning if 'x' is not involved as in the normal linear regression we have one step learning but here how to perform learning for the constant model. Any help wud be highly appreciated.


Thanks