I have been using multiple kernel learning (MKL) to train a classifier and got some exposure to the field. However, I am quite new to machine learning and I have only an intuitive understanding of the subject. I have recently been told that the MKL algorithm I am using might not be the best choice for what I am trying to do, which leads to my question:

How do you go about choosing the best algorithm for the job - say for example SimpleMKL, AverageMKL or EasyMKL? What is the intuition in the difference between these and what is the "ideal" application for each?

  • $\begingroup$ Can you define what MKL is? $\endgroup$ – Sycorax Jul 1 at 23:06
  • $\begingroup$ @Sycorax yes, I am mean multiple kernel learning (MKL) as described here: stackoverflow.com/a/62093700/12268981 $\endgroup$ – mm523 Jul 2 at 11:58
  • $\begingroup$ Ok, in that case you'll need to explain what SimpleMKL, AverageMKL and EasyMKL are and where they come from. Are they part of some software package? Is this terminology used in some academic literature? $\endgroup$ – Sycorax Jul 2 at 15:36
  • $\begingroup$ I have linked out the two original papers on SimpleMKL and EasyMKL in the question. Unfortunately, I struggle to understand them as my understanding is too intuitive for the depth they go into - hence why I was asking the question. AverageMKL is one of the options provided on MKLpy (github.com/IvanoLauriola/MKLpy1), but I could not find any formal literature on it. $\endgroup$ – mm523 Jul 2 at 17:34

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.