Yang Liu, Yegor Klochkov, Wenlong Chen
We lifted 7 functions out of this paper's own repositories and ran 7 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| chenw20/biasscore | canonical | 7 of 7 |
| Function | Status | Where it lives |
|---|---|---|
| Accuracy | Ran | chenw20/biasscore/postprocess_dp.py pointer only (licence: NONE) · get_code("36e09db3d9eee8a3") |
| Accuracy_a | Ran | chenw20/biasscore/postprocess_eo.py pointer only (licence: NONE) · get_code("6a9e4b82cd8dab2f") |
| Accuracy_y | Ran | chenw20/biasscore/postprocess_eo.py pointer only (licence: NONE) · get_code("552165348f78226c") |
| DP | Ran | chenw20/biasscore/postprocess_dp.py pointer only (licence: NONE) · get_code("31fdf66771eccfcd") |
| DP_sign | Ran | chenw20/biasscore/postprocess_dp.py pointer only (licence: NONE) · get_code("6647607303258f55") |
| EO | Ran | chenw20/biasscore/postprocess_eo.py pointer only (licence: NONE) · get_code("c89bb5cab361dd52") |
| compute_score | Ran | chenw20/biasscore/postprocess_dp.py pointer only (licence: NONE) · get_code("d3b2da90b6341d9e") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
We consider a binary classification problem under group fairness constraints, which can be one of Demographic Parity (DP), Equalized Opportunity (EOp), or Equalized Odds (EO). We propose an explicit characterization of Bayes optimal classifier under the fairness constraints, which turns out to be a simple modification rule of the unconstrained classifier. Namely, we introduce a novel instancelevel measure of bias, which we call bias score, and the modification rule is a simple linear rule on top of the finite amount of bias scores. Based on this characterization, we develop a post-hoc approach that allows us to adapt to fairness constraints while maintaining high accuracy. In the case of DP and EOp constraints, the modification rule is thresholding a single bias score, while in the case of EO constraints we are required to fit a linear modification rule with 2 parameters. The method can also be applied for composite group-fairness criteria, such as ones involving several sensitive attributes. We achieve competitive or better performance compared to both in-processing and post-processing methods across three datasets: Adult, COMPAS, and CelebA. Unlike most post-processing methods, we do not require access to sensitive attributes during the inference time.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2310.05725")
get_code_for_paper("2310.05725")
have("2310.05725")
Connect an agent — have() is free.