search

JuliaGaussianProcesses / KernelFunctions.jl

Julia package for kernel functions for machine learning
https://juliagaussianprocesses.github.io/KernelFunctions.jl/stable/
MIT License
267 stars 32 forks source link

Missing examples in docs #11

Open IsakFalk opened 4 years ago

IsakFalk commented 4 years ago

The examples in the docs is currently lacking, so it would be good to provide some examples.

Currently we have to fill out the following

I have a good knowledge of the first two, but shouldn't be very hard to fix the other ones as well, but will take some time.

theogf commented 4 years ago

Thanks for opening the issue. I already have drafts for kernel ridge regression, and I also made a simple SVM example. I need to find the references for the two last. It should be straight forward to make!

IsakFalk commented 4 years ago

No problem! Regarding the papers, the documentation references papers to use (I thought you were the one that added them, but maybe not!).

For deep kernel learning we have the original paper. We could probably just redo one of the experiments combining Flux with KernelFunctions and optimise it directly (conditioned on that the kernels work with Zygote / AutoDiff which I think but I'm not sure).

For the kernel selection part I can't really find a reference (at least not from AISTATS 2018 as referenced on the docs).

theogf commented 4 years ago

I finally found the AISTATS 2018 paper I was looking for! https://arxiv.org/abs/1706.02524