Nice explanation video! I have a few question regarding the RLS formulations at 5:59; If the output Y is not 1-dimensional, the matrix-inversion cannot be avoided right? And, a second question I have is the following: At 5:00 it is stated that one should only start calculating theta when R is invertible. How can this be done in the formulations at 5:59 as the recursion depends on the previous theta (meaning you are forced to calculate theta at every instance)? If there is no easy way of doing this with the formulations at 5:59, I will go with the one of 5:00.
Thank god for this resolution and crisp fps
Nice explanation video! I have a few question regarding the RLS formulations at 5:59; If the output Y is not 1-dimensional, the matrix-inversion cannot be avoided right? And, a second question I have is the following: At 5:00 it is stated that one should only start calculating theta when R is invertible. How can this be done in the formulations at 5:59 as the recursion depends on the previous theta (meaning you are forced to calculate theta at every instance)? If there is no easy way of doing this with the formulations at 5:59, I will go with the one of 5:00.
Thank you!