globus-labs / globus-labs/stopping-bayesian-optimization

Weight-Space Views?

Open
#1 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Jupyter Notebook
Stars
1
Forks
0
PR merge metrics
No merged PRs in 30d

Description

It seems like we can cast GPR as either a "weight-space" or a "function-space" distribution. See [GPR textbook chapter 2](http://www.gaussianprocess.org/gpml/chapters/RW2.pdf)

"Weight-space" might be what we need to make the "sample model parameters" implementation feasible.

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.