https://github.com/mnagired/fairness-sandbox
Repo for framework which assesses the viability of fairness-enhancing interventions.
https://github.com/mnagired/fairness-sandbox
fairness-ml machinelearning
Last synced: 3 months ago
JSON representation
Repo for framework which assesses the viability of fairness-enhancing interventions.
- Host: GitHub
- URL: https://github.com/mnagired/fairness-sandbox
- Owner: mnagired
- License: mit
- Created: 2021-06-07T15:44:30.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2022-08-08T02:45:07.000Z (about 3 years ago)
- Last Synced: 2025-03-23T18:52:21.110Z (7 months ago)
- Topics: fairness-ml, machinelearning
- Language: Jupyter Notebook
- Homepage:
- Size: 28.4 MB
- Stars: 0
- Watchers: 4
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Bias Injection Sandbox
A framework which assesses the effectiveness of fairness-enhancing interventions.
## Structure
`sandbox.ipynb`: main file to run the sandbox's functionalities
## Description
Our sandbox offers the following pipeline:
### Data
1. Upload Dataset
2. Choose existing dataset (e.g. Adult Income)
3. Generate Synthetic Dataset
### Model
1. Train any ml model of choice
### Bias Injection
Select one (or more) bias(es) to inject into the data from the following list:
1. Representation Bias (under-sampling subsets of the data)
2. Measurement Bias (adding noise)
3. Omitted Variable Bias
4. Label Noise Bias
5. Over-Sampling Bias
6. Under-Sampling Bias
### Fairness Intervention
Select one of the following interventions:
1. [Correlation Remover](https://fairlearn.org/v0.7.0/api_reference/fairlearn.preprocessing.html#module-fairlearn.preprocessing) (Pre-Processing)
2. [Exponentiated Gradient](https://fairlearn.org/v0.7.0/api_reference/fairlearn.reductions.html#fairlearn.reductions.ExponentiatedGradient) (In-Processing)
3. [Grid Search](https://fairlearn.org/v0.7.0/api_reference/fairlearn.reductions.html#fairlearn.reductions.GridSearch) (In-Processing)
4. [Threshold Optimizer](https://fairlearn.org/v0.7.0/api_reference/fairlearn.postprocessing.html) (Post-Processing)
### Fairness Visualization
After selecting a metric of your choice (e.g. accuracy, precision, roc_auc, etc), we output a plot which displays the effectiveness of the fairness intervention's ability to mitigate the bias you injected, with respect to the ground truth data.
## License
This project is licensed under the [MIT] License - see the LICENSE.md file for details