In supervised learning, we get a representer theorem by considering regularized losses of the following form:

enter image description here

In Kernel Density Estimation, we simply directly assume densities of the form

enter image description here

Could this be justified by for example regularized maximum likelihood? In other words, is there a representer theorem for unsupervised learning?



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.