The Impossibility Theorem of Machine Fairness – A Causal Perspective

07/12/2020
by   Kailash Karthik S, et al.
0

With the increasing pervasive use of machine learning in social and economic settings, there has been an interest in the notion of machine bias in the AI community. Models trained on historic data reflect the biases that exist in society and are propagated to the future through their decisions. A recent study conducted by ProPublica revealed that the COMPAS recidivism prediction tool was biased against the African-American community. There are three prominent metrics of fairness used in the community, and it has been statistically proved that it is impossible to satisfy them at the same time – which has led to ambiguity about the definition of fairness. In this report, causal perspective to the impossibility theorem of fairness is presented along with a causal goal for machine fairness.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
09/24/2018

Evaluating Fairness Metrics in the Presence of Dataset Bias

Data-driven algorithms play a large role in decision making across a var...
research
12/21/2021

A Pilot Study on Detecting Unfairness in Human Decisions With Machine Learning Algorithmic Bias Detection

Fairness in decision-making has been a long-standing issue in our societ...
research
04/16/2023

Fairness in AI and Its Long-Term Implications on Society

Successful deployment of artificial intelligence (AI) in various setting...
research
08/25/2020

Improving Fair Predictions Using Variational Inference In Causal Models

The importance of algorithmic fairness grows with the increasing impact ...
research
07/09/2020

Transparency Tools for Fairness in AI (Luskin)

We propose new tools for policy-makers to use when assessing and correct...
research
12/05/2019

Perfectly Parallel Fairness Certification of Neural Networks

Recently, there is growing concern that machine-learning models, which c...
research
05/09/2020

Cyberbullying Detection with Fairness Constraints

Cyberbullying is a widespread adverse phenomenon among online social int...

Please sign up or login with your details

Forgot password? Click here to reset