JuliaGaussianProcesses / JuliaGaussianProcesses/KernelFunctions.jl

Documenting caveats of `SimpleKernel`

Open
#520 7 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Julia
Stars
275
Forks
41
PR merge metrics
No merged PRs in 30d

Description

The docs recommend SimpleKernel for building ones own kernel using Distances.jl. They should probably here also note that not every PreMetric yields a positive-definite kernel.

In particular, as Theorem 1 of https://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Feragen_Geodesic_Exponential_Kernels_2015_CVPR_paper.pdf notes, the geodesic distance for any "non-flat" manifold does not yield a positive-definite kernel when used in a squared exponential kernel, which would mean e.g. Distances.SphericalAngle will not yield a PD kernel.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the documented SimpleKernel section, especially the guidance for kernel functions depending on a metric. Check the linked Theorem 1 and the Distances.SphericalAngle example, then update the documentation to state that some PreMetric choices do not produce positive-definite kernels. Done means the caveat and example are clearly included in that section.

Written by the indexing model from the issue text.

Assessment

Tech stack
julia
Domain
documentation, machine-learning
Issue type
Documentation
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Clearly specified
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.