I was learning about kernel mean embeddings. I was under the impression that the mean embedding was just the mean of a function with respect to a probability distribution. Then I saw somewhere that if you have a "one-hot" representation of a distribution the kernel embedding becomes a vector. Is this true? How can you use these mean embeddings like in SVMs or Ridge Regressions if they are vectors.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.