AI Case Study

Google Brain researchers develops a method to avoid machine learning model discrimination against legally protected attributes

Google Brain researchers propose a way in the post-processing of machine learning methods which determine threshold levels to control for potential discrimination on the groups assigned to those threshold levels. They demonstrate application of the technique using FICO credit scores.



Software And It Services

Project Overview

"We propose a simple, interpretable, and actionable framework for measuring and removing discrimination based on protected attributes. We examine various fairness measures in the context of FICO scores with the protected attribute of race. FICO scores are a proprietary classifier widely used in the United States to predict credit worthiness. FICO scores are complicated proprietary classifiers based on features, like number of bank accounts kept, that could interact with culture—and hence race—in unfair ways. ....we’ve proposed a methodology for measuring and preventing discrimination based on a set of sensitive attributes. Our framework not only helps to scrutinize predictors to discover possible concerns. We also show how to adjust a given predictor so as to strike a better tradeoff between classification accuracy and non-discrimination if need be.

At the heart of our approach is the idea that individuals who qualify for a desirable outcome should have an equal chance of being correctly classified for this outcome. In our fictional loan example, it means the rate of ‘low risk’ predictions among people who actually pay back their loan should not depend on a sensitive attribute like race or gender. We call this principle equality of opportunity in supervised learning."

Reported Results

"Assuming data about the predictor, target, and membership in the protected group are available, we show how to optimally adjust any learned predictor so as to remove discrimination according to our definition. Our framework also improves incentives by shifting the cost of poor classification from disadvantaged groups to the decision maker, who can respond by improving the classification accuracy."


"In this paper, we consider non-discrimination from the perspective of supervised learning, where the goal is to predict a true outcome Y from features X based on labeled training data,
while ensuring they are “non-discriminatory” with respect to a specified protected attribute A. As in the usual supervised learning setting, we assume that we have access to labeled training
data, in our case indicating also the protected attribute A.

We show that the Bayes optimal non-discriminating (according to our definition) classifier is the classifier derived from any Bayes optimal (not necessarily non-discriminating) regressor using our post-processing step. Moreover, we quantify the loss that follows from imposing our non-discrimination condition in case the score we start from deviates from Bayesian optimality. This result helps to justify the approach of deriving a fair classifier via post-processing rather than changing the original training process."


Legal And Compliance



The researchers explain the problem thusly: "an algorithm could, for example, be used to predict with high accuracy who will pay back their loan. Lenders might then use such a predictor as an aid in deciding who should receive a loan in the first place. Although machine learning aims to minimize the chance of a mistake, how do we prevent certain groups from experiencing a disproportionate share of these mistakes? Consider the case of a group that we have relatively little data on and whose characteristics differ from those of the general population in ways that are relevant to the prediction task. As prediction accuracy is generally correlated with the amount of data available for training, it is likely that incorrect predictions will be more common in this group. A predictor might, for example, end up flagging too many individuals in this group as ‘high risk of default’ even though they pay back their loan. When group membership coincides with a sensitive attribute, such as race, gender, disability, or religion, this situation can lead to unjust or prejudicial outcomes."



A sample of 301,536 TransUnion TransRisk scores from