I thought Radford Neal explains that, too. Google also papers on Deep GPs
I’ve seen some of the stuff on how going deep is equivalent to learning a GP with a non-stationary kernel. Is that the work you were referring to?
I thought Radford Neal explains that, too. Google also papers on Deep GPs
I’ve seen some of the stuff on how going deep is equivalent to learning a GP with a non-stationary kernel. Is that the work you were referring to?