Skip to content

Latest commit

 

History

History
65 lines (44 loc) · 3.55 KB

demographic_parity.rst

File metadata and controls

65 lines (44 loc) · 3.55 KB
Raw

Fairness/Demographic(Statistical) Parity Difference(Tabular)

Demographic Parity, also called Independence or Statistical Parity, is a well-known criteria to calculate fairness in machine learning. According to it, proportions of all segments of protected class (eg., gender) should receive equal rates of positive outcome. For example, the probability of getting admission to a college must be independent of gender. In simple terms, outcome must be independent of protected class.

This metric is computed as the difference between the rate of positive outcomes in unprivileged and privileged groups.

Sam Corbett-Davies, Emma Pierson, Avi Feller, Sharad Goel, and Aziz Huq. 2017.Algorithmic decision making and the cost of fairness. In Proceedings of KDD

Cynthia Dwork, Moritz Hardt, Toniann Pitassi, Omer Reingold, and Richard Zemel. 2012. Fairness through awareness. In Proceedings of the 3rd Innovations in Theoretical Computer Science Conference (ITCS). 214–226

Jon Kleinberg, Sendhil Mullainathan, and Manish Raghavan. 2017. Inherent Trade-Offs in the Fair Determination of Risk Scores. In Proceedings of ITCS.

Input Information

Property Notes
input Specify the dataset CSV file containing the data to analyze. file name of the current dataset will be set as default.
target_variable Specify the name of the column in the input CSV file to use as the target (label) variable.
privileged_variable Specify the name of the column in the input CSV file to use as the privileged variable (Class in the protected attribute with the majority is called privileged class).
unprivileged_variable Specify the name of the column in the input CSV file to use as the unprivileged variable (Class in the protected attribute with minority is called unprivileged class).
reweighing_weight Specify the name of the column in the input CSV file to use as reweighing weights that are assigned to individual samples.Default is NA. So each sample is given a unit weight.
fair_threshold Specify fairness threshold, between -1.0 & 1.0. Based on this value, model outputs whether outcome is "fair" or "unfair". Default value is 0.10. So, all outcomes between -0.1 and 0.1 are "fair".
num_samples Specify the number of samples to compute the Demographic parity, by default num_samples is "all", which means compute the Demographic parity of all the samples in the input file.
output Specify the name of the CSV file to output the Demographic parity(DPD) result.

Output Information

The result of this plugin is saved in the designated 'output' path as CSV file. The information on the columns of CSV file is as follows.

Fairness Plot Plot fairness check enables us to look how the fairness between privileged and unprivileged groups with respect to the fairness definition. If the bar plot reaches out of the green zone it means that for this subgroup fairness goal is not satisfied.
Demographic Parity The ideal value of this Demographic Parity metric is 0, value of Demographic Parity < 0 implies a higher benefit for the privileged group, value of Demographic Parity > 0 implies a higher benefit for the unprivileged group.