Lightning-AI / Lightning-AI/pytorch-lightning
Add Callback for Differential Privacy
Open
Nobody has claimed this yet.
callback
feature
- Dominant language
- Python
- Stars
- 31.4k
- Forks
- 3.8k
- Avg merge
- 6d 7h
- Merged PRs (30d)
- 6
Description
Why not include a callback that stops the training when the Differential Privacy budget has been spent?
See [20853](https://github.com/Lightning-AI/pytorch-lightning/pull/20853).
cc @lantiga @borda
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading the issue and the referenced pull request 20853 to understand the existing Differential Privacy integration and callback conventions. Define the stopping condition around exhaustion of the privacy budget, then verify that training halts when the budget is spent and that the behavior is covered by the relevant tests identified while reading the referenced work.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- ai, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 30/100