nick-jhlee/fair-manifold-pca

Fast and Efficient MMD-based Fair PCA via Optimization over Stiefel Manifold (AAAI 2022)

35
/ 100
Emerging

This project helps data scientists and machine learning engineers transform high-dimensional data into a smaller, more manageable form while actively reducing bias related to sensitive attributes like gender or race. It takes in your original dataset, along with sensitive group labels and optionally a downstream classification label, to produce a lower-dimensional representation that is fairer according to your chosen criteria. This tool is for practitioners building predictive models who need to ensure their data's dimensionality reduction steps do not inadvertently amplify or introduce unfairness.

No commits in the last 6 months.

Use this if you are performing dimensionality reduction (like PCA) on a dataset and need to explicitly mitigate bias against specific protected groups present in your data, ensuring fairness in downstream analyses or models.

Not ideal if your primary goal is simply general-purpose dimensionality reduction without a specific concern for fairness metrics or if you need a solution fully implemented in Python (this tool is predominantly in MATLAB).

fair-AI bias-mitigation data-preprocessing machine-learning-ethics predictive-modeling
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

11

Forks

3

Language

C++

License

MIT

Last pushed

Sep 27, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/nick-jhlee/fair-manifold-pca"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.