How can we find the low rank representation of unfamiliar data in Laplacian Eigenmaps? As in: in PCA you find a transformation matrix that you can apply to any data point you have and it will give you the representation of your data point in the different components. However, it seems in the laplacian eigenmaps that we directly solve for the low rank representation i.e y. Therefore if I have new data points I will have to do everything from scratch? ps: I feel that this has already been discussed in the video but I can't seem to find it again.
@alirezamogharabi87333 жыл бұрын
سلام آقای دکتر قدسی، نمیدونم این پیام بنده به شما میرسه یا خیر، اما هر بار که ویدیویی از شما میبینم از ته دل آرزو میکنم که یه روز بتونم سر کلاسهای حضرتعالی باشم. لیسانس و فوق لیسانس دانشگاه آزاد خوندم، اما امیدوارم دکترام واترپلو باشه.