On the (im)possibility of fairness

09/23/2016
by   Sorelle A. Friedler, et al.
0

What does it mean for an algorithm to be fair? Different papers use different notions of algorithmic fairness, and although these appear internally consistent, they also seem mutually incompatible. We present a mathematical setting in which the distinctions in previous papers can be made formal. In addition to characterizing the spaces of inputs (the "observed" space) and outputs (the "decision" space), we introduce the notion of a construct space: a space that captures unobservable, but meaningful variables for the prediction. We show that in order to prove desirable properties of the entire decision-making process, different mechanisms for fairness require different assumptions about the nature of the mapping from construct space to decision space. The results in this paper imply that future treatments of algorithmic fairness should more explicitly state assumptions about the relationship between constructs and observations.

READ FULL TEXT

page 3

page 9

research
06/18/2020

Algorithmic Decision Making with Conditional Fairness

Nowadays fairness issues have raised great concerns in decision-making s...
research
05/19/2020

Fair Inputs and Fair Outputs: The Incompatibility of Fairness in Privacy and Accuracy

Fairness concerns about algorithmic decision-making systems have been ma...
research
05/31/2017

Subjective fairness: Fairness is in the eye of the beholder

We analyze different notions of fairness in decision making when the und...
research
02/07/2023

From Utilitarian to Rawlsian Designs for Algorithmic Fairness

There is a lack of consensus within the literature as to how `fairness' ...
research
03/03/2023

FairShap: A Data Re-weighting Approach for Algorithmic Fairness based on Shapley Values

In this paper, we propose FairShap, a novel and interpretable pre-proces...
research
11/12/2017

Ensuring Liveness Properties of Distributed Systems (A Research Agenda)

Often fairness assumptions need to be made in order to establish livenes...
research
12/11/2019

Measurement and Fairness

We introduce the language of measurement modeling from the quantitative ...

Please sign up or login with your details

Forgot password? Click here to reset