CliMA / CliMA/EnsembleKalmanProcesses.jl

O3.7.5 Formulate and implement variant of EKI that scales linearly in output dimension and allows for sampling error correction

Open
#374 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Julia
Stars
125
Forks
24
Avg merge
1d 18h
Merged PRs (30d)
5

Description

# Purpose

To implement some changes that (1) make EKI (scalable in input dim) more scalable in output dimensions, and (2) make ETKI (scalable in output dim) more scalable in input dim

```[tasklist]
- [ ] https://github.com/CliMA/EnsembleKalmanProcesses.jl/issues/377
- [ ] EKI update linear algebra from online->offline to reduce online complexity
- [ ] Implement Localization implementations that is compatible with ETKI
```

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with linked issue #377 and the existing EKI and ETKI implementations; this issue does not identify files, tests, or entry points. Review the requested online-to-offline linear algebra and ETKI-compatible localization work. Done means the requested variants support the stated output and input scaling goals and sampling error correction.

Written by the indexing model from the issue text.

Assessment

Tech stack
julia
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.