For mixture of experts, what should I do if my data of experts are continuous and not linear?

I've read mixture of experts using linear experts but I can't find an example or a paper related to the use of continuous experts

20 June 2016 2,262 1 View